Chapter One: The Importance of Technological Innovation

Chapter One
Introduction
THE IMPORTANCE OF TECHNOLOGICAL INNOVATION
technological In many industries technological innovation is now the most important driver of
innovation competitive success. Firms in a wide range of industries rely on products developed
The act of within the past five years for almost one-third (or more) of their sales and profits. For
introducing a
new device,
example, at Johnson & Johnson, products developed within the last five years account
method, or for over 30 percent of sales, and sales from products developed within the past five
material for years at 3M have hit as high as 45 percent in recent years.
application to The increasing importance of innovation is due in part to the globalization of mar-
commercial or kets. Foreign competition has put pressure on firms to continuously innovate in order
practical
objectives.
to produce differentiated products and services. Introducing new products helps firms
protect their margins, while investing in process innovation helps firms lower their
costs. Advances in information technology also have played a role in speeding the
pace of innovation. Computer-aided design and computer-aided manufacturing have
made it easier and faster for firms to design and produce new products, while flex-
ible manufacturing technologies have made shorter production runs economical and
have reduced the importance of production economies of scale.1 These technologies
help firms develop and produce more product variants that closely meet the needs
of narrowly defined customer groups, thus achieving differentiation from competi-
tors. For example, in 2015, Toyota offered 21 different passenger vehicle lines under
the Toyota brand (e.g., Camry, Prius, Highlander, and Tundra). Within each of the
vehicle lines, Toyota also offered several different models (e.g., Camry L, Camry LE,
and Camry SE) with different features and at different price points. In total, Toyota
offered 167 car models ranging in price from $14,845 (for the Yaris three-door lift-
back) to $80,115 (for the Land Cruiser), and seating anywhere from three passengers
(e.g., Tacoma Regular Cab truck) to eight passengers (Sienna Minivan). On top of
this, Toyota also produced a range of luxury vehicles under its Lexus brand. Similarly,
Samsung introduced 52 unique smartphones in 2014 alone. Companies can use broad
portfolios of product models to help ensure they can penetrate almost every conceiv-
able market niche. While producing multiple product variations used to be expensive
1
2 Chapter 1 Introduction
and time-consuming, flexible manufacturing technologies now enable firms to seam-

lessly transition from producing one product model to the next, adjusting production
schedules with real-time information on demand. Firms further reduce production
costs by using common components in many of the models.
As firms such as Toyota, Samsung, and others adopt these new technologies and
increase their pace of innovation, they raise the bar for competitors, triggering an
industrywide shift to shortened development cycles and more rapid new product
introductions. The net results are greater market segmentation and rapid product obso-
lescence.2 Product life cycles (the time between a product’s introduction and its with-
drawal from the market or replacement by a next-generation product) have become
as short as 4 to 12 months for software, 12 to 24 months for computer hardware and
consumer electronics, and 18 to 36 months for large home appliances.3 This spurs
firms to focus increasingly on innovation as a strategic imperative—a firm that does
not innovate quickly finds its margins diminishing as its products become obsolete.
THE IMPACT OF TECHNOLOGICAL INNOVATION ON SOCIETY

If the push for innovation has raised the competitive bar for industries, arguably mak-
ing success just that much more complicated for organizations, its net effect on society
is more clearly positive. Innovation enables a wider range of goods and services to be
delivered to people worldwide. It has made the production of food and other neces-
sities more efficient, yielded medical treatments that improve health conditions, and
enabled people to travel to and communicate with almost every part of the world. To
get a real sense of the magnitude of the effect of technological innovation on society,
look at Figure 1.1, which shows a timeline of some of the most important technologi-
cal innovations developed over the last 200 years. Imagine how different life would be
without these innovations!
The aggregate impact of technological innovation can be observed by looking at
gross gross domestic product (GDP). The gross domestic product of an economy is its
domestic total annual output, measured by final purchase price. Figure 1.2 shows the average
product (GDP) GDP per capita (that is, GDP divided by the population) for the world, developed
The total annual countries, and developing countries from 1969 to 2014. The figures have been con-
output of an
economy as verted into U.S. dollars and adjusted for inflation. As shown in the figure, the average
measured by its world GDP per capita has risen steadily since 1969. In a series of studies of economic
final purchase growth conducted at the National Bureau of Economic Research, economists showed
price. that the historic rate of economic growth in GDP could not be accounted for entirely
by growth in labor and capital inputs. Economist Robert Merton Solow argued that
this unaccounted-for residual growth represented technological change: Technologi-
cal innovation increased the amount of output achievable from a given quantity of
labor and capital. This explanation was not immediately accepted; many researchers
attempted to explain the residual away in terms of measurement error, inaccurate price
deflation, or labor improvement. But in each case the additional variables were unable
to eliminate this residual growth component. A consensus gradually emerged that the
Chapter 1 Introduction 3
FIGURE 1.1 residual did in fact capture techno-

1800 - 1800—Electric battery
Timeline logical change. Solow received a
- 1804—Steam locomotive
of Some of Nobel Prize for his work in 1981,
- 1807—Internal combustion engine
The Most and the residual became known as
- 1809—Telegraph
Important - 1817—Bicycle the Solow Residual.4 While GDP
Technological
1820 - 1821—Dynamo has its shortcomings as a measure
Innovations In
The Last 200
- 1824—Braille writing system of standard of living, it does relate
- 1828—Hot blast furnace
Years very directly to the amount of
- 1831—Electric generator
- 1836—Five-shot revolver goods consumers can purchase.
1840 - 1841—Bunsen battery (voltaic cell) Thus, to the extent that goods
- 1842—Sulfuric ether-based anesthesia improve quality of life, we can
- 1846—Hydraulic crane ascribe some beneficial impact of
- 1850—Petroleum refining technological innovation.
- 1856—Aniline dyes
Sometimes technological inno-
1860 - 1862—Gatling gun
externalities vation results in negative extern-
- 1867—Typewriter
Costs (or benefits)
- 1876—Telephone alities. Production technologies may
that are borne create pollution that is harmful to
- 1877—Phonograph
(or reaped) by
individuals
- 1878—Incandescent lightbulb the surrounding communities; agri-
other than those 1880 - 1885—Light steel skyscrapers cultural and fishing technologies
responsible for - 1886—Internal combustion automobile can result in erosion, elimination
- 1887—Pneumatic tire
creating them. of natural habitats, and depletion
Thus, if a - 1892—Electric stove
- 1895—X-ray machine of ocean stocks; medical technolo-
business emits
pollutants in a 1900 - 1902—Air conditioner (electric) gies can result in unanticipated
community, it - 1903—Wright biplane consequences such as antibiotic-
imposes a nega- - 1906—Electric vacuum cleaner resistant strains of bacteria or
tive externality - 1910—Electric washing machine moral dilemmas regarding the use
on the community - 1914—Rocket
of genetic modification. However,
members; if a 1920 - 1921—Insulin (extracted)
business builds a technology is, in its purest essence,
- 1927—Television
park in a commu- - 1928—Penicillin knowledge—knowledge to solve
nity, it creates a - 1936—First programmable computer our problems and pursue our
positive external- - 1939—Atom fission goals.5 Technological innovation is
ity for community 1940 - 1942—Aqua lung thus the creation of new knowledge
members. - 1943—Nuclear reactor that is applied to practical prob-
- 1947—Transistor
lems. Sometimes this knowledge
- 1957—Satellite
- 1958—Integrated circuit is applied to problems hastily,
1960 - 1967—Portable handheld calculator without full consideration of the
- 1969—ARPANET (precursor to Internet) consequences and alternatives, but
- 1971—Microprocessor overall it will probably serve us
- 1973—Mobile (portable cellular) phone better to have more knowledge
- 1976—Supercomputer
than less.
1980 - 1981—Space shuttle (reusable)
- 1987—Disposable contact lenses
- 1989—High-definition television
- 1990—World Wide Web protocol
- 1996—Wireless Internet
2000 - 2003—Map of human genome
FIGURE 1.2
Gross $50,000
Domestic
Product per $45,000
Capita, 1969–
$40,000
2014 (in Real
2010 $US $35,000
Billions)
Source: USDA
$30,000
Economic Research
Service, International $25,000
Macroeconomic
Dataset (http://www.
$20,000
ers.usda.gov, accessed
August 17, 2015)
$15,000
$10,000
$5,000
$0
20 9
20 5
20 3
20 7
20 1
69
99
95
89
85
13
93
83
79
97
75
87
73
77
11
0
0
0
0
0
91
81
71
20
20
19
19
19
19
19
19
19
19
19
19
19
19
19
19
19
19
World Developed Countries Developing Countries
INNOVATION BY INDUSTRY: THE IMPORTANCE OF STRATEGY

As will be shown in Chapter Two, the majority of effort and money invested in tech-
nological innovation comes from industrial firms. However, in the frenetic race to
innovate, many firms charge headlong into new product development without clear
strategies or well-developed processes for choosing and managing projects. Such firms
often initiate more projects than they can effectively support, choose projects that are
a poor fit with the firm’s resources and objectives, and suffer long development cycles
and high project failure rates as a consequence (see the accompanying Research Brief
for a recent study of the length of new product development cycles). While innova-
tion is popularly depicted as a freewheeling process that is unconstrained by rules and
plans, study after study has revealed that successful innovators have clearly defined
innovation strategies and management processes.6
The Innovation Funnel

Most innovative ideas do not become successful new products. Many studies suggest
that only one out of several thousand ideas results in a successful new product: Many
projects do not result in technically feasible products and, of those that do, many fail
to earn a commercial return. According a 2012 study by the Product Development and
Management Association, only about one in nine projects that are initiated are success-
ful, and of those that make it to the point of being launched to the market, only about
half earn a profit.7 Furthermore, many ideas are sifted through and abandoned before
Research Brief How Long Does New Product

Development Take?a
In a large-scale survey administered by the Prod- significantly longer, clocking in at 57 weeks. The
uct Development and Management Association development of radical products or technologies
(PDMA), researchers examined the length of time took the longest, averaging 82 weeks. The study
it took firms to develop a new product from initial also found that on average, for more innovative
concept to market introduction. The study divided and radical projects, firms reported significantly
new product development projects into catego- shorter cycle times than those reported in the pre-
ries representing their degree of innovativeness: vious PDMA surveys conducted in 1995 and 2004.
“radical” projects, “more innovative” projects, and a
Adapted from Markham, SK, and Lee, H. “Product
“incremental” projects. On average, incremental Development and Management Association’s 2012
projects took only 33 weeks from concept to mar- comparative performance assessment study,” Journal
ket introduction. More innovative projects took of Product Innovation Management 30 (2013), issue 3:
408–429.
a project is even formally initiated. According to one study that combined data from
prior studies of innovation success rates with data on patents, venture capital fund-
ing, and surveys, it takes about 3,000 raw ideas to produce one significantly new and
successful commercial product.8 The pharmaceutical industry demonstrates this
well—only one out of every 5,000 compounds makes it to the pharmacist’s shelf, and
only one-third of those will be successful enough to recoup their R&D costs.9 Further-
more, most studies indicate that it costs at least $1.5 billion and a decade of research to
bring a new Food and Drug Administration (FDA)-approved pharmaceutical product
to market! 10 The innovation process is thus often conceived of as a funnel, with many
potential new product ideas going in the wide end, but very few making it through the
development process (see Figure 1.3).
FIGURE 1.3
The New Prod-
uct Develop-
ment Funnel in
Pharmaceuticals
5,000 125
2-3 drugs tested 1 drug
Compounds Leads Rx
Discovery & Preclinical Clinical Trials Approval

3–6 years 6–7 years ½–2 years
The Strategic Management of Technological Innovation

Improving a firm’s innovation success rate requires a well-crafted strategy. A firm’s
innovation projects should align with its resources and objectives, leveraging its core
competencies and helping it achieve its strategic intent. A firm’s organizational struc-
ture and control systems should encourage the generation of innovative ideas while
also ensuring efficient implementation. A firm’s new product development process
should maximize the likelihood of projects being both technically and commercially
successful. To achieve these things, a firm needs (a) an in-depth understanding of the
dynamics of innovation, (b) a well-crafted innovation strategy, and (c) well-designed
processes for implementing the innovation strategy. We will cover each of these in
turn (see Figure 1.4).
In Part One, we will cover the foundations of technological innovation, gaining an
in-depth understanding of how and why innovation occurs in an industry, and why
some innovations rise to dominate others. First, we will look at the sources of innova-
tion in Chapter Two. We will address questions such as: Where do great ideas come
from? How can firms harness the power of individual creativity? What role do cus-
tomers, government organizations, universities, and alliance networks play in creating
innovation? In this chapter we will first explore the role of creativity in the generation of
novel and useful ideas. We then look at various sources of innovation, including the role
of individual inventors, firms, publicly sponsored research, and collaborative networks.
In Chapter Three, we will review models of types of innovation (such as radical ver-
sus incremental and architectural versus modular) and patterns of innovation (including
s-curves of technology performance and diffusion, and technology cycles). We will
address questions such as: Why are some innovations much harder to create and imple-
ment than others? Why do innovations often diffuse slowly even when they appear to
offer a great advantage? What factors influence the rate at which a technology tends to
improve over time? Familiarity with these types and patterns of innovation will help
us distinguish how one project is different from another and the underlying factors that
shape the project’s likelihood of technical or commercial success.
In Chapter Four, we will turn to the particularly interesting dynamics that emerge in
industries characterized by increasing returns, where strong pressures to adopt a single
dominant design can result in standards battles and winner-take-all markets. We will
address questions such as: Why do some industries choose a single dominant standard
rather than enabling multiple standards to coexist? What makes one technological
innovation rise to dominate all others, even when other seemingly superior technolo-
gies are offered? How can a firm avoid being locked out? Is there anything a firm can
do to influence the likelihood of its technology becoming the dominant design?
In Chapter Five, we will discuss the impact of entry timing, including first-mover
advantages, first-mover disadvantages, and the factors that will determine the firm’s
optimal entry strategy. This chapter will address such questions as: What are the
advantages and disadvantages of being first to market, early but not first, and late?
What determines the optimal timing of entry for a new innovation? This chapter
reveals a number of consistent patterns in how timing of entry impacts innovation suc-
cess, and it outlines what factors will influence a firm’s optimal timing of entry, thus
beginning the transition from understanding the dynamics of technological innovation
to formulating technology strategy.
FIGURE 1.4
The Strategic Management of Technological Innovation
Part 1: Industry Dynamics of

Technological Innovation
Chapter 2 Chapter 3 Chapter 4 Chapter 5

Sources of Types and Patterns Standards Battles Timing of Entry
Innovation of Innovation and Design
Dominance
Part 2: Formulating Technological

Innovation Strategy
Chapter 6
Defining the Organization’s
Strategic Direction
Chapter 7 Chapter 8 Chapter 9

Choosing Innovation Collaboration Protecting Innovation
Projects Strategies
Part 3: Implementing Technological

Innovation Strategy
Chapter 10 Chapter 11 Chapter 12 Chapter 13

Organizing for Managing the New Managing New Crafting a
Innovation Product Development Product Deployment
Process Development Teams Strategy
Feedback
In Part Two, we will turn to formulating technological innovation strategy. Chapter

Six reviews the basic strategic analysis tools managers can use to assess the firm’s
current position and define its strategic direction for the future. This chapter will
address such questions as: What are the firm’s sources of sustainable competitive
advantage? Where in the firm’s value chain do its strengths and weaknesses lie? What
are the firm’s core competencies, and how should it leverage and build upon them?
What is the firm’s strategic intent—that is, where does the firm want to be 10 years
from now? Only once the firm has thoroughly appraised where it is currently can it
formulate a coherent technological innovation strategy for the future.
In Chapter Seven, we will examine a variety of methods of choosing innovation
projects. These include quantitative methods such as discounted cash flow and options
valuation techniques, qualitative methods such as screening questions and balancing
the research and development portfolio, as well as methods that combine qualitative
and quantitative approaches such as conjoint analysis and data envelopment analysis.
Each of these methods has its advantages and disadvantages, leading many firms to
use a multiple-method approach to choosing innovation projects.
In Chapter Eight, we will examine collaboration strategies for innovation. This
chapter addresses questions such as: Should the firm partner on a particular project or
go solo? How does the firm decide which activities to do in-house and which to access
through collaborative arrangements? If the firm chooses to work with a partner, how
should the partnership be structured? How does the firm choose and monitor partners?
We will begin by looking at the reasons a firm might choose to go solo versus working
with a partner. We then will look at the pros and cons of various partnering methods,
including joint ventures, alliances, licensing, outsourcing, and participating in col-
laborative research organizations. The chapter also reviews the factors that should
influence partner selection and monitoring.
In Chapter Nine, we will address the options the firm has for appropriating the
returns to its innovation efforts. We will look at the mechanics of patents, copyright,
trademarks, and trade secrets. We will also address such questions as: Are there ever
times when it would benefit the firm to not protect its technological innovation so
vigorously? How does a firm decide between a wholly proprietary, wholly open, or
partially open strategy for protecting its innovation? When will open strategies have
advantages over wholly proprietary strategies? This chapter examines the range of
protection options available to the firm, and the complex series of trade-offs a firm
must consider in its protection strategy.
In Part Three, we will turn to implementing the technological innovation strategy.
This begins in Chapter Ten with an examination of how the organization’s size and
structure influence its overall rate of innovativeness. The chapter addresses such ques-
tions as: Do bigger firms outperform smaller firms at innovation? How do formaliza-
tion, standardization, and centralization impact the likelihood of generating innovative
ideas and the organization’s ability to implement those ideas quickly and efficiently?
Is it possible to achieve creativity and flexibility at the same time as efficiency and
reliability? How do multinational firms decide where to perform their development
activities? How do multinational firms coordinate their development activities toward
a common goal when the activities occur in multiple countries? This chapter examines
how organizations can balance the benefits and trade-offs of flexibility, economies of
scale, standardization, centralization, and tapping local market knowledge.
In Chapter Eleven, we will review a series of “best practices” that have been identi-
fied in managing the new product development process. This includes such questions as:
Should new product development processes be performed sequentially or in parallel?
What are the advantages and disadvantages of using project champions? What are
the benefits and risks of involving customers and/or suppliers in the development
process? What tools can the firm use to improve the effectiveness and efficiency of
its new product development processes? How does the firm assess whether its new
product development process is successful? This chapter provides an extensive review
of methods that have been developed to improve the management of new product
development projects and to measure their performance.
Chapter Twelve builds on the previous chapter by illuminating how team composi-
tion and structure will influence project outcomes. This chapter addresses questions
such as: How big should teams be? What are the advantages and disadvantages of
choosing highly diverse team members? Do teams need to be collocated? When
should teams be full-time and/or permanent? What type of team leader and manage-
ment practices should be used for the team? This chapter provides detailed guidelines
for constructing new product development teams that are matched to the type of new
product development project under way.
Finally, in Chapter Thirteen, we will look at innovation deployment strategies.
This chapter will address such questions as: How do we accelerate the adoption of the
technological innovation? How do we decide whether to use licensing or OEM agree-
ments? Does it make more sense to use penetration pricing or a market-skimming
price? When should we sell direct versus using intermediaries? What strategies can
the firm use to encourage distributors and complementary goods providers to sup-
port the innovation? What are the advantages and disadvantages of major marketing
methods? This chapter complements traditional marketing, distribution, and pricing
courses by looking at how a deployment strategy can be crafted that especially targets
the needs of a new technological innovation.
Summary 1. Technological innovation is now often the single most important competitive
of driver in many industries. Many firms receive more than one-third of their sales
and profits from products developed within the past five years.
Chapter
2. The increasing importance of innovation has been driven largely by the global-
ization of markets and the advent of advanced technologies that enable more
rapid product design and allow shorter production runs to be economically
feasible.
3. Technological innovation has a number of important effects on society, includ-
ing fostering increased GDP, enabling greater communication and mobility, and
improving medical treatments.
4. Technological innovation may also pose some negative externalities, including
pollution, resource depletion, and other unintended consequences of technological
change.
5. While government plays a significant role in innovation, industry provides the

majority of R&D funds that are ultimately applied to technological innovation.
6. Successful innovation requires an in-depth understanding of the dynamics of
innovation, a well-crafted innovation strategy, and well-developed processes for
implementing the innovation strategy.
Discussion 1. Why is innovation so important for firms to compete in many industries?

Questions 2. What are some advantages of technological innovation? Disadvantages?
3. Why do you think so many innovation projects fail to generate an economic return?
Suggested Classics
Further Arrow, K. J., “Economic welfare and the allocation of resources for inventions,”
Reading in The Rate and Direction of Inventive Activity: Economic and Social Factors, ed.
R. Nelson (Princeton, NJ: Princeton University Press, 1962), pp. 609–25.
Mansfield, E., “Contributions of R and D to economic growth in the United States,”
Science CLXXV (1972), pp. 477–86.
Schumpeter, J. A., The Theory of Economic Development (1911; English translation,
Cambridge, MA: Harvard University Press, 1936).
Stalk,G. and Hout, T.M., “Competing Against Time: How Time-Based Competition Is
Reshaping Global Markets” (New York: Free Press, 1990).
Recent Work
Ahlstrom, D., “Innovation and growth: How business contributes to society,” Acad-
emy of Management Perspectives, (2010) August, pp. 10–23.
Baumol, W. J., The Free Market Innovation Machine: Analyzing the Growth Miracle
of Capitalism (Princeton, NJ: Princeton University Press, 2002).
Editors, “The top 25 innovations of the last 25 years,” Popular Science (2012),
November 15th. (www.popsci.com)
Friedman, T. L., The World Is Flat: A Brief History of the Twenty-First Century (New
York: Farrar, Straus and Giroux, 2006).
Schilling, M.A. 2015. Towards dynamic efficiency: Innovation and its implications for
antitrust. Forthcoming in Antitrust Bulletin.
Endnotes 1. J. P. Womack, D. T. Jones, and D. Roos, The Machine That Changed the World (New York:
Rawson Associates, 1990).
2. W. Qualls, R. W. Olshavsky, and R. E. Michaels, “Shortening of the PLC—an Empirical Test,”
Journal of Marketing 45 (1981), pp. 76–80.
3. M. A. Schilling and C. E. Vasco, “Product and Process Technological Change and the Adoption of
Modular Organizational Forms,” in Winning Strategies in a Deconstructing World, eds. R. Bresser,
M. Hitt, R. Nixon, and D. Heuskel (Sussex, England: John Wiley & Sons, 2000), pp. 25–50.
CHAPTER 3. CONCEPTS AND DEFINITIONS FOR MEASURING BUSINESS INNOVATION │ 67
Chapter 3. Concepts and definitions for measuring business innovation
This chapter provides a set of definitions to guide statistical surveys of innovation within
the Business sector, including a taxonomy for different types of innovation. The definitions
within this chapter also help characterise business enterprises in relation to their innovations
and their activities in pursuit of innovation. The aim of his chap er s defini ions and
complementary guidance is to facilitate the collection and reporting of comparable data
on innovation and related activities for firms in different countries and industries and for
firms of different sizes and structures, ranging from small single-product firms to large
multinational firms responsible for a wide range of products (goods or services). The
chapter concludes with recommendations on the use of definitions in surveys.
OSLO MANUAL 2018 © OECD/EUROPEAN UNION 2018

68 │ CHAPTER 3. CONCEPTS AND DEFINITIONS FOR MEASURING BUSINESS INNOVATION
3.1. Introduction
3.1. Based on the concepts presented in Chapter 2, this chapter provides a set of
definitions to guide statistical surveys of innovation within the Business sector. As innovation
is a pervasive, heterogeneous and multi-faceted phenomenon, clear and concise definitions
for innovation and related concepts are required for accurate measurement and interpretation
of business innovation activities and to establish a common standard that serves the needs
of the producers and users of innovation statistics.
3.2. The definitions given in this chapter facilitate the collection and reporting of comparable
data on innovation and related activities for firms in different countries and industries and
for firms of different sizes and structures, ranging from small single-product firms to large
multinational firms that produce a wide range of products, including services.
3.3. Section 3.2 contains the main definitions for measuring innovation in the Business
enterprise sector. Section 3.3 develops various taxonomies of business innovation including
by type, and by novelty and impacts. Changes that are not innovations are described in
section 3.4. Section 3.5 categorises firms according to their innovation status. Section 3.6
concludes with recommendations on the use of definitions in surveys.
3.2. Innovation in the Business enterprise sector
3.2.1. Definition of innovation activities and innovation

3.4. As discussed in Chapter 2, the term inno ation can be sed in different conte ts
to refer to either a process or an outcome. To avoid confusion, this manual uses the term
inno ation acti ities to refer to the process hile the term inno ation is limited to o tcomes.
3.5. The basic definition of (business) innovation activities is as follows:
Innovation activities include all developmental, financial and commercial activities
undertaken by a firm that are intended to result in an innovation for the firm.
3.6. Innovation activities can result in an innovation (defined below), be ongoing,
postponed or abandoned. Follow-on activities as defined in subsection 4.5.3 are generally
outside the scope of innovation activity.
3.7. The organisation of innovation activities varies greatly between firms. Some firms
manage their innovation activities through well-defined innovation projects or programmes
with dedicated budgets, for which an innovation represents an intermediate or final milestone.
Other firms primarily integrate their innovation activities into regular business operations
and work to make continuous improvements to their products and business processes, while
other firms primarily engage in innovation activities on an ad hoc basis. All methods of
organising innovation activities are within the scope of the definitions and recommendations
in this chapter. Additional details on the definition, categorisation and measurement of
innovation activities are provided in Chapter 4.
3.8. This chapter focuses on the concept of innovation and provides summary
definitions for innovation and for different types of innovation. Each definition is followed
by additional details on the interpretation of the definition.
3.9. The basic definition of a business innovation is as follows:
A business innovation is a new or improved product or business process (or combination
thereof) that differs significantly from the firm's previous products or business processes
and that has been introduced on the market or brought into use by the firm.

3.10. As introduced in Chapter 2, a product is a good or service (or combination thereof).

Business processes include all core activities by the firm to produce products and all
ancillary or supporting activities.
3.11. A product is introduced when it is made available for use by its intended users. A
b siness process is introd ced hen it is bro ght into act al se in the firm s operations.
The act of introduction is defined as implementation and is the point in time when a
significantly different product or business process is first made available for use. Firms will
often make further adjustments to an innovation after its implementation (see Chapter 4),
for instance to the characteristics of a new service. Some of these can be sufficiently
different to count as an additional innovation.
3.12. The minimum requirement for an innovation is that the product or business process
must have one or more characteristics that are significantly different from those contained
in the products or business processes previously offered by or used by the firm. These
characteristics must be relevant to the firm or to external users. For example, the firm may
expect the new or improved characteristics of a product (or business process) to increase
utility for users or to enhance its own competitive position in the market. Relevant characteristics
are described below for product innovations and business process innovations.
3.13. An innovation can also result from a series of minor improvements made during
the observation period, provided that the sum of these minor improvements results in a
significant difference in the final product or business process.
3.14. The requirement for significantly different characteristics applies to product and
business process innovations that a firm develops itself and innovations first developed by
other firms, organisations or individuals, with little or no additional modification. Therefore,
the definition of innovation also includes diffusion.
3.15. The adoption of a new or improved product or business process by a firm that is
part of an enterprise group is an innovation, even if the new or improved product or business
process has previously been introduced on the market or brought into use by other firms within
the same enterprise group. For instance, the adoption, by a subsidiary firm, of a new business
process that was developed and brought into use by the parent firm is an innovation for the
subsidiary firm. However, the adoption of a new or improved product or business process that
was already in use in a different section or division of the same firm is not an innovation.
3.16. The concept of a significant difference excludes minor changes or enhancements.
However, the boundary between a change that is an innovation and one that is not an
innovation is na oidabl s bjecti e beca se it is relati e to each firm s conte t, capabilities
and requirements. For example, an improvement to an online service could represent a
minor change for a large firm in a research and experimental development (R&D)-intensive
industry but be a significant difference for a small firm in a less R&D-intensive industry.
3.17. The definition does not require an innovation to be a commercial, financial or
strategic success at the time of measurement. A product innovation can fail commercially
or a business process innovation may require more time to meet its objectives.
3.18. The definition of an innovation does not require it to have a positive value for
society, or a positive benefit for the firm. In the former case, an innovation can lead to a
significant boost in the financial performance of the firm while providing fewer benefits to
consumers than other offerings from the same firm or its competitors. An innovation can
also result in safety, health or environmental problems. Conversely, an innovation does not
necessarily improve the market position or financial performance of the firm when their
users benefit from it. For example, an innovation can improve the utility for users without
increasing a firm's sales, market share or net earnings.

3.2.2. Division of innovation effort and responsibilities

3.19. The division of labour that underpins economic specialisation also applies to
innovation activities, as a majority of firms are unlikely to possess all of the necessary
capabilities and property rights to develop an innovation. Many innovations are based on
purchasing, imitating or modifying products, business process equipment, or business
methods that are already in use by other firms or organisations. Consequently many firms
do not develop all of the concepts, prototypes or designs that underpin their innovations
and multiple firms can derive similar innovations from a single concept or technology. Nor
do firms implement all of the concepts or prototypes they develop, for example when a firm
only licenses an invention to other firms. These relationships and how they result in
different types of innovations are discussed in detail in Chapter 6.
3.20. Innovations that have been developed in full or in part elsewhere, or in partnership
with third parties are not necessarily less valuable; they may only signal a higher degree of
specialisation. Data collection should encourage respondents to report all innovations,
including those that were not primarily developed by their own firm.
3.3. Taxonomies of innovation
3.21. Innovation changes the characteristics of one or more products or business processes
and consequently common usage describes innovation in terms of its purpose or object. For
example, managers may refer to their firm s ser ice inno ations or to a deli er s stem
innovation. Information on the object of an innovation is useful for assessing the purpose
of the innovation, its general characteristics, its potential impacts on the firm, and the types
of innovation activities that are relevant to its development and implementation.
3.3.1. Innovation types by object: Product and business process innovations

3.22. There are two major types of innovation by object: innovations that change the
firm s prod cts (prod ct inno ations), and inno ations that change the firm s business
processes (business process innovations).
3.23. Product innovations are divided into two main types, while business process
innovations are divided into six broad types (see below). A single innovation can involve
combinations of different types of product and business process innovations. Consequently,
the typology of innovation types by object is not a classification of mutually exclusive
categories. Furthermore, a firm can introduce more than one type of innovation over the
observation period for data collection. It is therefore recommended to collect information
on multiple types of innovations on the assumption that the responses can refer either to
different innovations or to innovations that combine two or more innovation types.
Product innovation
3.24. The term product is defined in the System of National Accounts and encompasses
both goods and services. Products are the economic output of production activities. They
can be exchanged and used as inputs in the production of other goods and services, as final
consumption by households or governments, or for investment, as in the case of financial
products (EC et al., 2009).
A product innovation is a new or improved good or service that differs significantly
from he firm s pre io s goods or ser ices and ha has been in rod ced on he marke .

3.25. Product innovations must provide significant improvements to one or more

characteristics or performance specifications. This includes the addition of new functions,
or improvements to existing functions or user utility. Relevant functional characteristics
include quality, technical specifications, reliability, durability, economic efficiency during
use, affordability, convenience, usability, and user friendliness. Product innovations do not
need to improve all functions or performance specifications. An improvement to or addition
of a new function can also be combined with a loss of other functions or a decline in some
performance specifications.
3.26. Relevant characteristics can include financial attributes such as affordability and
financial convenience. Examples of innovations with financial characteristics that provide
benefits to users include dynamic toll pricing to ease traffic congestion, the introduction of
a new product line that uses less expensive materials and is consequently offered at lower
cost, and a service for automatic payment of a taxi ride after the ride has taken place.
3.27. An additional characteristic of both goods and services that can influence usability
or utility is product design. New designs or improved design features can influence the
appearance or look of a prod ct and conseq entl enhance the ser s tilit , for instance
through a substantial design change that creates a positive emotional response. However,
minor design changes are unlikely to lead to goods or services that differ significantly from
previous ones (see below).
3.28. A product innovation must be made available to potential users, but this does not
require the innovation to generate sales. Limiting product innovations to those with sales
would exclude product innovations that fail to meet established or expected demand or
where sales require a longer observation period to materialise. In addition, this would
exclude digital products that are offered at no cost to users, with revenue obtained from
advertising, monetising user information, or through other methods.
3.29. Product innovations can use new knowledge or technologies, or be based on new
uses or combinations of existing knowledge or technologies.
Types of products
3.30. Product innovations can involve two generic types of products: goods and services.
These product types have been introduced in Chapter 2 and are defined below drawing on
the System of National Accounts (SNA) (EC et al., 2009).
Goods include tangible objects and some knowledge-capturing products (see
below) over which ownership rights can be established and whose ownership can
be transferred through market transactions.
Services are intangible activities that are produced and consumed simultaneously
and that change the conditions (e.g. physical, psychological, etc.) of users. The
engagement of users through their time, availability, attention, transmission of
information, or effort is often a necessary condition that leads to the co-production
of services by users and the firm. The attributes or experience of a service can
therefore depend on the input of users. Services can also include some knowledge-
capturing products (see below).
3.31. As noted in Chapter 2, the dividing line between goods and services can sometimes
be difficult to establish and some products can have characteristics of both. A company can
sell goods to its customers or rent their use as a service, as is often the case for durable
consumer goods and for assets for business production. Firms can also bundle ancillary
services such as service contracts or insurance with their goods.

3.32. Knowledge-capturing products (as identified in the SNA) can have the characteristics
of either a good or service and concern the provision, storage, safekeeping, communication
and dissemination of digital information that users can access repeatedly. These products
can be stored on physical objects and infrastructure, such as electronic media or the Cloud.
An example is when access to digital products such as music, films and books is provided
on demand to consumers for a fee. Knowledge-capturing products are similar to a good if
consumers can share or sell them to others after purchase, but they are similar to a service
if the cons mer s rights are limited b a license that restricts sharing or selling. Digital
technologies, through reducing the cost of copying and exchanging information to a
negligible amount, have contributed to the proliferation of knowledge-capturing products.
3.33. At a minimum, it is recommended to collect data on both goods and services.
Surveys should specifically refer to services to ensure that the questions are relevant to
respondents from service sector firms. Where possible, data should be collected on
knowledge-capturing products, especially those of a digital nature, to support research on
the prevalence of these products and the factors that influence their development.
Business process innovation

3.34. All business functions can be the object of innovation activity. The term business
process includes the core business function of producing goods and services and supporting
functions such as distribution and logistics, marketing, sales and after-sales services;
information and communication technology (ICT) services to the firm, administrative and
management functions, engineering and related technical services to the firm, and product
and business process development. Business processes can be considered as services for
which the firm itself is the customer. Business processes can be delivered in-house or
procured from external sources.
A business process innovation is a new or improved business process for one or
more business functions that differs significan l from he firm s pre io s b siness
processes and that has been brought into use in the firm.
3.35. The relevant characteristics of an improved business function are related to those
for an improved product, in particular services that can be delivered to business customers.
Examples include greater efficacy, resource efficiency, reliability and resilience, affordability,
convenience and usability for those involved in the business process, either external or
internal to the firm.
3.36. Both new and improved business processes can be motivated by goals to implement
business strategies, reduce costs, improve product quality or working conditions, or to meet
regulatory requirements. A business process innovation can involve improvements to one
or more aspects of a single business function or to combinations of different business
functions. They can involve the adoption by the firm of new or improved business services that
are delivered by external contractors, for instance accounting or human resources systems.
3.37. Business process innovations are implemented when they are brought into use by
the firm in its internal or outward-facing operations. The implementation of a business
process innovation can require several steps, from initial development, pilot testing in
a single business function, to implementation across all relevant business functions.
Implementation occ rs hen the b siness process is sed on an ongoing basis in the firm s
operations. This can occur shortly after pilot testing.
3.38. Digital technologies and practices are pervasive across business processes. They
are used to codify processes and procedures, add functions to existing processes and enable

the sale of processes as services. The implementation of business process innovations is

therefore often tied to the adoption and modification of digital technologies.
Types of business processes

3.39. Business process innovations concern the different functions of a firm. Management
research has produced several lists of business functions that differ by the definition of core
functions (activities that produce income) and supporting business functions, and by how
different activities are grouped (Brown, 2008). Business functions have proved useful for
the st d of global al e chains, for e ample in Canada s S rvey of Innovation and
Business Strategy (SIBS) and the European Survey on International Sourcing of Business
Functions (see Chapter 7).
Table 3.1. Functional categories for identifying the type of business process innovations
Short term Details and subcategories

1. Production of goods Activities that transform inputs into goods or services, including engineering and related
or services technical testing, analysis and certification activities to support production.
2. Distribution and This function includes:
logistics a) transportation and service delivery
b) warehousing
c) order processing.
3. Marketing and sales This function includes:
a) marketing methods including advertising (product promotion and placement, packaging of
products), direct marketing (telemarketing), exhibitions and fairs, market research and other
activities to develop new markets
b) pricing strategies and methods
c) sales and after-sales activities, including help desks other customer support and customer
relationship activities.
4. Information and The maintenance and provision of information and communication systems, including:
communication a) hardware and software
systems b) data processing and database
c) maintenance and repair
d) web-hosting and other computer-related information activities.
These functions can be provided in a separate division or in divisions responsible for other functions.
5. Administration and This function includes:
management a) strategic and general business management (cross-functional decision-making), including
organising work responsibilities
b) corporate governance (legal, planning and public relations)
c) accounting, bookkeeping, auditing, payments and other financial or insurance activities
d) human resources management (training and education, staff recruitment, workplace
organisation, provision of temporary personnel, payroll management, health and medical support)
e) procurement
f) managing external relationships with suppliers, alliances, etc.
6. Product and Activities to scope, identify, develop, or adapt products or a firm's business processes. This
business process function can be undertaken in a systematic fashion or on an ad hoc basis, and be conducted
development within the firm or obtained from external sources. Responsibility for these activities can lie
within a separate division or in divisions responsible for other functions, e.g. production of
goods or services.
Source: Adapted from Brown (2008), B siness processes and b siness f nctions: A ne a of looking at
emplo ment , www.bls.gov/mlr/2008/12/art3full.pdf and Eurostat (2018), Glossary of Statistical Terms,
http://ec.europa.eu/eurostat/statistics-explained/index.php/Glossary:Business_functions.
3.40. Table 3.1 provides a list of six main business functions based on the relevant
management and statistical literature that may be the object of innovation. The function
production of goods and services constitutes the core function of a firm, whereas the

other five functions comprise ancillary activities to support production and bring products
to the market. Firms can develop business process innovations that target one or more
functions. For example, the implementation of an online ordering system could represent
an innovation in to the distribution and logistics business functions. The short descriptions
of each business function, followed by the detailed description, are recommended for use
in data collection. The list is sufficiently brief for use in surveys and provides moderate
comparability with the definitions of process, organisational, and marketing innovations in
the third edition of the Oslo Manual. More detailed applications of this taxonomy can
improve comparability with the results of innovation surveys that followed the third edition
of this manual. The new categories also cover areas that were not identified in the third
edition, such as changes in financing (item 5c) and changes in functions dedicated to
product or process development (item 6).
3.41. The latter captures business process innovations in the business function dedicated
to the development of products and other business processes of the firm. There was no
equivalent type of business process in earlier editions of this manual. Examples of
innovations in this function include the use of new gene editing technologies to develop
either existing or new plant varieties or pharmaceuticals and the application of data mining
analysis to large databases to identify potential market development opportunities. Other
examples for an innovation in this category include the adoption of new methodologies
such as design thinking, co-creation, rapid prototyping or high-throughput screening. An
innovation of this type may just seek to introduce incremental modifications that do not
qualify as innovations e.g. to be able to cater to different c stomers needs or may seek
to bring about product or business process innovations. However, there is no guarantee that
such innovations will ultimately materialise.
3.42. For data collection, some functions can be combined into a single item or disaggregated.
For example, functions 1 and 6 could be combined into a single function that includes both
production activities and the development of products and business processes. Functions 3
and 5 could be further disaggregated to facilitate comparison with the definitions of
organisational and marketing innovation in the third edition of the manual (see next section
for details).
Comparison of innovation types with the previous edition of the Oslo Manual
3.43. Table 3.2 compares the types of product and business process innovations used in
this manual with the definitions used in the third edition of the Oslo Manual.
3.44. Two types of marketing innovation that are included in the third edition of the Oslo
Manual (adoption of methods for product placement and product promotion or pricing) are
not listed in the short description of the six business functions in Table 3.1, but these are
included in the detailed descriptions. In addition, this manual assigns innovations involving
the design of products under product innovation, whereas the third edition included these
under marketing innovation. The change is due to the close relationship between design
activities and the development of product characteristics for both goods and services.
However, changes in the design of packaging remain under marketing.
3.45. There is a good match between the fourth edition and the third edition s definitions
for two types of business process innovations, namely the production of goods and services
and for distribution and logistics. The third edition s s bcategor of ancillary services is
divided in this edition between information and communication systems on the one hand
and administration and management on the other, with the latter including activities that
are listed in the third edition under organisational innovation.

Table 3.2. Comparing types of innovation in the current and previous Oslo Manual editions
Fourth edition, 2018 (OM4) compared to third edition, 2005 (OM3)
OM3 OM3 subcomponents OM41 Differences

Product Goods Goods Inclusion of product design
Services Services characteristics, which were included
Goods and services include under marketing innovation in OM3.
knowledge-capturing
products, and combinations
thereof.
Includes the design
characteristics of goods and
services.
Process Production Production Ancillary services in OM3 moved to
Delivery and logistics Distribution and logistics administration and management.
Ancillary services, including Information and
purchasing, accounting and communication systems
ICT services
Organisational Business practices Administration and Organisational innovations in OM3 are
Workplace organisation management under administration and management
(distribution of subcategories a, b and f in this edition of
responsibilities) the manual.
External relations Ancillary services in administration and
management (subcategories c, d, and e)
were included under process innovation
in OM3.
Marketing Design of products Marketing, sales and after- Marketing innovations in OM3 are
Product placement and sales support included under subcategories a and b in
packaging this manual.
Product promotion Innovations in sales, after-sales
Pricing services, and other customer support
functions were not included in OM3.
Innovations related to product design
are included under product innovation in
this manual.
N/A N/A Product and business process Not explicitly considered in OM3 , most
development likely reported as Process innovation.
1. Additional granularity is possible by disaggregating the detailed descriptions in Table 3.1.
3.46. Empirical research has shown that business managers can find it difficult to
differentiate between organisational and process innovations. Organisational innovations
in this manual are therefore subsumed under one type of business process (administration
and management) that includes activities that can involve what previously was described
as organisational innovation, such as strategic management (business practices and external
relations in the third edition) and human resource management (workplace organisation in
the third edition).
3.47. The third edition of the manual supported the construction of a category of product
or process innovators only that excluded firms that were only organisational or marketing
innovators. This category can be approximated using this man al s categor of prod ct
innovation plus three business process categories: (i) production of goods or services;
(ii) distribution and logistics; and (iii) information and communication systems. The
approximation is not perfect because of differences between the third and current edition in
the classification of different types of product design, purchasing and accounting services.

3.48. Previous innovation surveys that followed the third edition of this manual collected
data on multiple types of innovation. For example, the European Community Innovation
Survey (CIS) collected data on two types of product innovations, three types of process
innovations, four types of organisational innovations and four types of marketing
innovations. This data can be reanalysed to approximate the innovation categories in
Table 3.1, thus minimising the impact of a break in series. However, there are several
exceptions where surveys based on the third edition cannot replicate the categories of this
manual, due to a lack of coverage of several administrative and management functions
(e.g. corporate governance), financing, after-sales services, and the business function of
product and business process development.
Combinations of several innovation types by object

3.49. Many innovations are bundled, presenting characteristics that span more than one
t pe (O Brien et al., 2015; Frenz and Lambert, 2012; OECD, 2013). This is due to the
complementarity between different types of innovations. Some possible combinations of
innovation types are as follows:
A business process innovation can significantly improve the quality of a product,
resulting in a joint business process and product innovation.
A product innovation can require a supporting business process innovation. This is
particularly common for service innovations. For example, a new online function
for selling information products is both a business process innovation (requiring
ICT and web development) and a service innovation for potential users. If it creates
a new sales channel for the first time, it can also be a marketing innovation.
Product and business process innovation can be closely intertwined, especially
when the process is not distinguishable from the product. This applies particularly
to services for which production, delivery and consumption occur simultaneously.
Changes by the firm to non-economic outputs of production processes, such as
carbon or NOx emissions from energy generation, are due to innovations in business
processes, but firms can choose to include emission changes in the product
description if there is market demand. In this example, low emission energy can be
a business process innovation and a product innovation.
3.50. The object approach discussed in Chapter 10 can help obtain information on the
incidence of different types of bundled innovations.
Business model innovations

3.51. A business model includes all core business processes such as the production,
logistical, marketing and co-operative arrangements in use as well as the main products
that a firm sells, currently or in the future, to achieve its strategic goals and objectives. A
firm can use a single business model or several business models at the same time, for
instance for different product lines or markets. The innovation management literature notes
that successful business models combine a method for better meeting the needs of users
relative to what competitors can deliver and a profit formula for earning income from
delivering utility to customers (Johnson, Christensen and Kagermann, 2008).
3.52. There is no single, recognised definition of a business model innovation, which can
vary from partial b siness model inno ations that onl affect either a firm s prod cts or
business functions, to comprehensive business model innovations that involve both products

and business functions. In many cases it is difficult to distinguish partial business model
innovations from product and business process innovations.
3.53. Comprehensive business model innovations are of greater interest because they can
have substantial effects on supply chains and economic production, transforming markets
and potentially creating new ones. They can influence how a firm creates utility for users
(product innovation) and how products are produced, brought to market, or priced (business
process innovations).
3.54. There are three types of comprehensive business model innovations in existing
firms: (i) a firm extends its business to include completely new types of products and
markets that require new business processes to deliver; (ii) a firm ceases its previous
activities and enters into new types of products and markets that require new business
processes; and (iii) a firm changes the business model for its existing products, for example
it switches to a digital model with new business processes for production and delivery and
the product changes from a tangible good to a knowledge-capturing service.
3.55. It is not recommended to directly collect data on business model innovation as a
distinct, stand-alone category through innovation surveys because of the difficulty in
differentiating partial business model innovations from other types. However, the occurrence
of comprehensive business model innovations could be estimated through analysis (see
Chapter 11) that combines information on the types of innovations introduced by a firm with
other questions on innovation objectives, including a question on the objective of establishing
a new business model (see Chapter 8). Identifying the third type of comprehensive business
model innovation could require dedicated questions on changes to existing products.
3.3.2. Types of innovation according to novelty and impacts

3.56. The basic requirement for an innovation is that it must be significantly different
from the firm s pre io s prod cts or b siness processes. As significantly different is
subjective and will vary according to the firm s capabilities and conte t, the interpretation
and comparability of innovation statistics can benefit from additional data on the significance
of innovations in terms of their novelty or economic impacts. Some forms of novelty, such
as disruptive or radical innovations, and some types of economic impacts are difficult to
identify within the limited observation period recommended for innovation surveys.
Alternative measures of novelty, innovativeness and economic impacts that are suitable
for survey observation periods include:
whether an inno ation is ne to the firm onl , ne to the firm s market, or ne to
the world
the firm s expectation of the potential to transform the market in which it operates
the firm s expectation of the potential to improve its competitiveness.
3.57. The first and most idel sed approach is to determine the no elt of a firm s
innovations (or at least one of its innovations) in comparison with the state of the art in the
market or industry in which the firm operates. A firm can serve a single market (if it only
offers one type of product) or several markets (if it offers different types of products). A
market can be geographically restricted (if a firm only serves customers in specific regions)
or it can be global. A firm can sell its products directly on local, regional, national or
international markets or through the use of intermediaries. Innovation can also create new
markets, which could allow the innovative firm to benefit from monopoly prices for a
certain period of time.

3.58. It is recommended to ask respondents if their firm has one or more product
innovations or business process innovations that are a market novelty (i.e. a first to their
market innovation). The interpretation of market novelty must be combined with information
on the geographical area served by the firm. A local or regional market novelty could be
based on imitating what is already available in other geographical markets, whereas a
world-first innovation will be a market leader.
3.59. Respondents can find it difficult to estimate if they have a world-first product
innovation, unless the innovation is based on one or more patented inventions that
underwent rigorous screening to establish global novelty. A world-first product innovation
implies a qualitatively greater degree of novelty than a new-to-market innovation.
3.60. Firms that first develop innovations are often drivers of follow-on innovation
within an industry. New ideas and knowledge often originate from these firms, but the
economic impact of their innovations will usually depend on the adoption (or imitation) of
their innovations by other firms. Information on the degree of novelty can be used to
identify the developers, adopters and imitators of innovations, to examine patterns of
diffusion, and to identify market leaders and followers.
3.61. The novelty of business process innovations in comparison to what is already in
use by other firms can be difficult for respondents to determine due to the importance of
secrecy and confidentiality for protecting business processes. However, evidence from
cognitive testing suggests that many managers are able to assess the novelty of process
innovations in their market, particularly for their most important business process innovations.
F rthermore, a don t kno response can pro ide al able information on the extent to
which secrecy is used in specific industries or types of firms.
3.62. The second option on the potential for an innovation to transform (or create) a
market can provide a possible indicator for the incidence of a radical or a disruptive
innovation. Radical innovations are considered to transform the status quo, while a disruptive
innovation takes root in simple applications in a niche market and then diffuses throughout
the market, eventually displacing established competitors (Christensen, 1997). Although
managers may be able to estimate the potential of an innovation to transform a market,
radical and disruptive innovations are likely to be very rare and therefore innovation
surveys may be a poor instrument for their detection. Relevant questions should be limited
to a single, most important innovation (see Chapter 10).
3.63. The third option on the effect of inno ations on the firm s competiti eness can be
assessed for product innovations through the observed change in sales over the observation
period (see Chapter 4) or by asking directly about future expectations of the effect of
innovations on competitiveness (see Chapter 7).
3.4. Changes that are not innovations
3.64. This section discusses changes that are either not an innovation or which can only
be considered an innovation if specific conditions are met. The basic principles are those
introduced earlier in section 3.2, namely that an innovation must have been implemented
and m st be significantl different from the firm s pre io s prod cts or b siness processes.
3.65. Routine changes or updates do not by themselves represent product innovations.
This includes software updates that only identify and remove coding errors and seasonal
changes in clothing fashions.

3.66. Simple capital replacement or extension is not an innovation. This includes the
purchase of identical models of installed equipment or minor extensions and updates to
existing equipment or software. New equipment or extensions must be new to the firm and
involve a significant improvement in specifications.
3.67. Product introductions that only involve minor aesthetic changes, such as a change
in colour or a minor change in shape, do not meet the requirement for a significant
difference and are therefore not product innovations.
3.68. Firms engaged in custom production make single and often complex goods or
services for sale on the market (e.g. computer games, films) or according to customer
orders (e.g. buildings, production plants, logistic systems, machinery, consulting reports).
Unless the one-off item displays significantly different attributes from products that the
firm has previously made, it is not a product innovation. It is not a business process
innovation unless developing the one-off item required the firm to develop and use
significantly different or enhanced capabilities. However, the first use of customised
production can be a business process innovation.
3.69. An advertised concept, prototype or model of a product that does not yet exist
is in general not a product innovation because it does not meet the implementation
requirement, even if customers can pre-order or make advance payments for the concept,
such as a product concept funded by crowdsourcing. The concept can fail or take
considerably longer than expected before it is available for use.
3.70. It may be more difficult to decide whether implementation has taken place in the
case of new knowledge products that have been sold to other parties. While the seller has
brought a new product to the market, the buyer may hold on from using it in their business
processes or taking it to their own markets. Such information may not be known to the
knowledge provider that is the subject of measurement and has to decide on whether to
report an innovation. If the knowledge product meets the novelty and significance requirements
to be considered a product innovation, a knowledge product can be considered to pass the
implementation test if it has been sold in the market by a firm to another party or parties.
3.71. The outputs of creative and professional service firms, such as reports for clients,
books, or films are not by default an innovation for the firms that develop them. For
example, a report by a consulting firm that summarises the results of a design project
without major novelty elements conducted under contract for a client is not a product
inno ation for the cons lting firm. The report s role in inno ation for the buying firm
depends on hether or not its res lts are sed in the client firm s inno ation acti ities.
However, the consulting firm could be credited with an innovation if it implemented new
business processes as part of conducting the project for its client, or if the blueprints or
designs that are sold on the market meet the innovation requirements of novelty and
significance. These phenomena are considered in more detail in Chapters 4 and 6.
3.72. Actions by retail, wholesale, transport and storage, and personal service firms to
extend the range of products handled or offered to customers are only an innovation if
the extension requires significant changes by the firm to its business processes. A fruit
importer or wholesaler who adds a new variety of fruit for sale to retailers is not engaged
in innovation unless the extension requires a major change to business processes such as
developing a new supply chain or the purchase of novel refrigerating equipment (e.g. to
permit the delivery of fresh produce that was not previously possible).
3.73. The activities of newly created firms (most of which are service firms) present a
potential source of confusion with respect to the basic definition of an innovation because

for a period of time a new firm will have no previous products or business processes for
comparison. In this case, the comparison group is what is available in the relevant market.
A product of a new firm is an innovation if it differs significantly from products available
in its markets. Likewise, a business process of a new firm is a process innovation if it differs
significantly from the business processes used by its competitors. However, respondents
from new firms may view all of their products or business processes as innovations.
Consequently it may be necessary to provide separate results for newly created firms such
as start-ups. In addition, it would be worthwhile for specialised surveys of start-up firms to
experiment with measuring product and business process novelty.
3.74. In the absence of further qualification, mergers or the acquisition of other firms
are not business process innovations in their own right. Mergers and acquisitions can drive
business process innovations, however, if the firm develops or adopts a new business
process as a result of the merger or for the purpose of improving the success of the merger
or acquisition.
3.75. Ceasing to use a business process, ceasing to outsource a business process, or
withdrawing a product from the market are not innovations. However, the first
implementation of business processes to determine when an activity should cease could
meet the requirements for an innovation.
3.76. A change due to externally determined factor prices is unlikely to represent an
innovation. For example, an innovation does not occur when the same model of a mobile phone
is constructed and sold at a lower price simply because the price of a video processor chip falls.
3.77. The formulation of a new corporate or managerial strategy is not an innovation
if it is not implemented. Furthermore, a change in a business process is not an innovation
if it is already in use in an identical form in other divisions of the firm.
3.5. Innovation and business profiling
3.5.1. Innovative and innovation-active firms

3.78. The innovation status of a firm is defined on the basis of its engagement in
innovation activities and its introduction of one or more innovations over the observation
period of a data collection exercise. As discussed in Chapter 9, the recommended observation
period can vary between one and three years.
3.79. During the observation period, any given innovation activity of the firm can:
Result in an innovation. The innovation activity can consequently cease during
the observation period after implementation or it could still be ongoing if it is
undertaken for other innovation projects.
Be ongoing without an innovation. Work can still be in progress and proceeding
according to plan, or delayed due to various reasons, such as technical difficulties
or a shortage of expertise or finance.
Be aborted, discontinued, or put on hold, for instance when activities to develop an
innovation are stopped before implementation.
3.80. These three outcomes apply to the wide array of innovation activities and projects
within a firm. The combination of data on the incidence of innovation and innovation
activity (innovation status) produces four possible categories for the innovative status of a
firm, as shown in Table 3.3.

Table 3.3. Innovative and innovation-active firms
The firm has innovation activities in the observation period

Yes No
The firm has at least one Yes The firm has one or more innovations and is It might occur if all work to introduce
innovation in the therefore an innovative firm. Innovation an innovation was conducted before
observation period. activities can be ongoing, put on hold, the observation period.
completed, or abandoned.
No The firm is innovation-active, but has not The firm is not engaged in innovation
introduced an innovation, although it might do activities and has not introduced any
so in the future. innovations in the observation
period.
3.81. The combinations in Table 3.3 result in three core definitions that apply to firms:
An innovative firm reports one or more innovations within the observation period.
This applies equally to a firm that is individually or jointly responsible for an innovation.
A non-innovative firm reports no innovations within the observation period.
An innovation-active firm is engaged at some time during the observation period
in one or more activities to develop or implement new or improved products or
business processes for an intended use. Both innovative and non-innovative firms
can be innovation-active during an observation period.
3.82. The fourth category of an innovative firm with no innovation activities during the
observation period is very rare. It would for example occur if a firm undertook all innovation
activities except implementation before the observation period and the implementation
required no additional resources. It may also occur if an innovation results from generic
business activities that were not explicitly aimed at introducing an innovation.
3.83. It is important for measurement practices to account for the dynamic relationship
between innovation viewed as a process (innovation activities) and as an outcome. The
length of the observation period will also directly influence the distribution of firms across
the four categories in Table 3.3. In industries with short development times and long
product life cycles, a short observation period could result in a low percentage of innovative
and innovation-active firms. In industries with long development times, a short observation
period could result in a high share of innovation-active firms combined with a low share of
innovative firms that report at least one innovation. Chapter 9 provides further discussion
of the effect of the observation period length on innovation status.
3.6. Use of innovation definitions in data collection
3.84. Innovation is a subjective construct with the potential for measurement to give
di erging res lts, depending on the respondent s perspecti e, beliefs and conte t (Galindo-
Rueda and Van Cruysen, 2016). To ensure statistical quality and comparability, the definitions
used in surveys and other data collection methods must therefore capture the intended
meaning of the definitions in this manual, while taking into account differences in language
and the vocabulary used and understood by potential respondents.
3.6.1. Use of the term innovation in surveys

3.85. An innovation survey can be designed to never use the term innovation in order
to a oid conflicts bet een the formal definition of an inno ation and each respondent s

own understanding. This could result in more objective responses and reduce issues of
comparability across industries or countries. An example is the Australian Business
Characteristics Survey, which replaces the term innovation with a description of all types
of innovations. For instance, the 2013 survey (based on the third edition of the Oslo
Manual) asks respondents here did this b siness so rce ideas and information for the
development or introduction of new goods, services, processes or methods? . This also
illustrates an important disadvantage of avoiding the use of innovation : it can require
listing all types of innovations in multiple questions. However, the adoption in this manual
of only two major categories of innovations, products and business processes, will improve
the ability of data collection exercises to avoid the term innovation while ensuring some
economy of language.
3.6.2. Innovation profiles

3.86. The minimum definition of an innovative firm is a poor indicator for comparing
innovation across industries, firm size classes or countries because it does not capture
ariations in the no elt of inno ations or each firm s capacit to de elop inno ations.
Information on firms innovation status can be combined with other information on
innovation novelty, innovation activities (see Chapter 4), or the division of innovation effort
(see Chapter 5) to produce indicators for the novelty of innovations and the innovation
capability of each firm. These indicators can be aggregated to produce innovation profiles
for firms by industry, firm size category or country. When combined with outcome data
(see Chapter 11), profiles can be used to explore the contribution of innovation to firm
performance and the utility for users of the innovation.
3.6.3. Priorities for data collection about innovations

3.87. It is recommended to collect data on the following topics of relevance to research
on innovation status and innovation profiles (see Chapter 11).
3.88. Data on each main innovation type by object (product and business process) can be
collected through a single question for each type, but it is useful for interpretation to include
additional questions on the two types of product innovations and the six types of business
process innovations. This will result in considerably more detailed information on the
innovations of each firm and permit replication of the generic innovation types (i.e. product
or process innovations) defined in the third edition of this manual.
3.89. The collection of data on innovation characteristics and novelty is recommended in
order to create innovation profiles that classify firms according to the characteristics of their
innovations and innovation efforts. Relevant questions for the construction of profiles include:
The different levels of innovation novelty, as per subsection 3.3.2.
The characteristics of product innovations, including design, as per subsection 3.3.1.
The role of third parties in developing and implementing innovations, as per
subsection 3.2.2 and Chapter 5.
The existence of ongoing or discontinued innovation activities, as per
subsection 3.5.1.
3.90. The concept of novelty is applicable to both product and business process
innovations, but questions on novelty are likely to be easier for managers to answer for
product innovations.

PART I
Defining and measuring R&D:

General guidance
FRASCATI MANUAL 2015 © OECD 2015

Frascati Manual 2015
Guidelines for Collecting and Reporting Data on Research
and Experimental Development
© OECD 2015
Chapter 2
Concepts and definitions

for identifying R&D
This chapter provides the definition of research and experimental

development (R&D) and of its components, basic research, applied research
and experimental development. These definitions are essentially unchanged
from those in previous editions of the manual. Where there are differences,
they reflect changes in culture and the use of language. To provide guidance
on what is and what is not an R&D activity, five criteria are provided which
require the activity to be novel, creative, uncertain in it outcome, systematic
and transferable and/or reproducible. Since the last edition, the treatment
of R&D expenditure in the System of National Accounts (SNA) has changed
from an expense to a capital investment. As a result, the language of this
manual, and of the SNA, is closer and there are additional requirements
for measurements of financial flows. While the manual has always
applied to all scientific disciplines, there is more emphasis on the social
sciences, humanities and the arts, in addition to the natural sciences and
engineering. Measuring R&D activities through surveys, administrative
data, or interviews raises questions about boundaries and what is and
what is not included and this chapter provides examples to help answer
those questions. The manual is used to interpret R&D data as part of policy
development and evaluation, but the focus of this chapter is on definitions
for measurement purposes.
43
I-2. CONCEPTS AND DEFINITIONS FOR IDENTIFYING R&D
2.1. Introduction
2.1 The Frascati Manual has provided the definition of research and
experimental development (R&D) and of its components, basic research, applied
research and experimental development, for more than half a century, and the
definitions have stood the test of time. The definitions in this chapter do not
differ in substance from those in previous editions. However, there is recognition
of cultural change in the definition of R&D and of the use of language in the
definition of experimental development.
2.2 Since the previous edition of this manual, the System of National
Accounts (SNA) has changed the treatment of expenditure on R&D from an
expense to a capital investment leading to a capital stock of knowledge created
as a result of R&D. The SNA 2008 (EC et al., 2009) draws on this manual for the
definition of R&D. A consequence of becoming a more integral part of the SNA is
the use of its language in this manual. Such usage will be noted when it occurs.
2.3 R&D is found in the social sciences, humanities and the arts as well as in
the natural sciences and engineering. This manual gives greater emphasis than
past editions to the social sciences, humanities and the arts. This requires no
changes in the definitions and conventions, but it does require greater attention
to the boundaries that define what is and what is not R&D. Also, countries using
this manual are at different stages of economic development, and this chapter
tries to accommodate the differing needs.
2.4 The chapter provides definitions of R&D and its components, together
with a set of criteria for identifying R&D. Examples of R&D, boundaries and
exclusions are provided to illustrate how the definitions are applied. This is a
statistical manual, and its fundamental purpose is to provide guidance for
the measurement of R&D activities using various means of data-gathering
from surveys, interviews and administrative sources. The manual is also used
for interpreting R&D data as part of the development, implementation and
evaluation of policy. However, users should note that the focus of this chapter is
on definitions for measurement purposes.
2.2. Definition of research and experimental development (R&D)

2.5 Research and experimental development (R&D) comprise creative
and systematic work undertaken in order to increase the stock of knowledge –
including knowledge of humankind, culture and society – and to devise new
applications of available knowledge.
44 FRASCATI MANUAL 2015 © OECD 2015

2.6 A set of common features identifies R&D activities, even if these are
carried out by different performers. R&D activities may be aimed at achieving
either specific or general objectives. R&D is always aimed at new findings,
based on original concepts (and their interpretation) or hypotheses. It is largely
uncertain about its final outcome (or at least about the quantity of time and
resources needed to achieve it), it is planned for and budgeted (even when
carried out by individuals), and it is aimed at producing results that could be
either freely transferred or traded in a marketplace. For an activity to be an R&D
activity, it must satisfy five core criteria.
2.7 The activity must be:
● novel
● creative
● uncertain
● systematic
● transferable and/or reproducible.
2.8 All five criteria are to be met, at least in principle, every time an
R&D activity is undertaken whether on a continuous or occasional basis. The
definition of R&D just given is consistent with the definition of R&D used in
the previous editions of the Frascati Manual and covers the same range of
activities.
2.9 The term R&D covers three types of activity: basic research, applied
research and experimental development. Basic research is experimental
or theoretical work undertaken primarily to acquire new knowledge of the
underlying foundations of phenomena and observable facts, without any
particular application or use in view. Applied research is original investigation
undertaken in order to acquire new knowledge. It is, however, directed primarily
towards a specific, practical aim or objective. Experimental development is
systematic work, drawing on knowledge gained from research and practical
experience and producing additional knowledge, which is directed to producing
new products or processes or to improving existing products or processes. These
three types of R&D are discussed further in Section 2.5.
2.10 This manual follows the System of National Accounts convention
in which “product” refers to a good or a service (EC et al., 2009: para. 2.36).
Throughout this manual, “process” refers to the transformation of inputs to
outputs and their delivery or to organisational structures or practices.
2.11 The order in which the three types of R&D activity appear is not
meant to suggest that basic research leads to applied research and then to
experimental development. There are many flows of information and knowledge
in the R&D system. Experimental development can inform basic research, and
there is no reason why basic research cannot lead directly to new products or
processes.

45
2.3. R&D activities and projects

2.12 An “R&D activity” is the sum of actions deliberately undertaken
by R&D performers in order to generate new knowledge. In most cases, R&D
activities can be grouped to form “R&D projects”. Each R&D project consists of a
set of R&D activities, is organised and managed for a specific purpose, and has its
own objectives and expected outcomes, even at the lowest level of formal activity.
The concept of an R&D project, while useful for understanding how R&D is done,
is not likely to be applied in the same way in all the sectors used in this manual.
2.4. The five criteria for identifying R&D

2.13 For an activity to be classified as an R&D activity, five core criteria have
to be jointly satisfied. A set of examples, which is by no means exhaustive, is
used to illustrate how the five criteria can be effectively applied to identify R&D
activities as well as specific R&D projects.
To be aimed at new findings (novel)

2.14 New knowledge is an expected objective of an R&D project, but it has
to be adapted to different contexts. For example, research projects in universities
are expected to pursue entirely new advancements in knowledge, and the same
can be said for projects designed and managed by research institutes.
2.15 In the Business enterprise sector (Frascati Manual sectors are defined
in Chapter 3), the potential novelty of R&D projects has to be assessed by
comparison with the existing stock of knowledge in the industry. The R&D activity
within the project must result in findings that are new to the business and not
already in use in the industry. Excluded from R&D are activities undertaken
to copy, imitate or reverse engineer as a means of gaining knowledge, as this
knowledge is not novel.
2.16 Novelty could result from a project to reproduce an existing result
that finds potential discrepancies. An experimental development project aimed
at creating knowledge in support of the development of new concepts and ideas
related to the design of new products or processes should be included in R&D.
As R&D is the formal creation of knowledge, including knowledge embodied in
products and processes, the measurement focus is on the new knowledge, not
on the new or significantly improved products or processes resulting from the
application of the knowledge. An example of R&D could be the integration of the
“maintenance manual” of a very complex system (like a passenger aircraft) with
additional material emerging from practical experience in ordinary maintenance
and properly codified, so long as this was done as part of an R&D project. Another
example is systematic testing to provide documentation of the potential use
of a chemical reaction that has already been adopted in production processes
(an existing technology) in order to achieve a new molecule, which has been
considered an improbable outcome by the scientific literature.

To be based on original, not obvious, concepts

and hypotheses (creative)
2.17 An R&D project must have as an objective new concepts or ideas
that improve on existing knowledge. This excludes from R&D any routine
change to products or processes and, therefore, a human input is inherent to
creativity in R&D. As a result, an R&D project requires the contribution of a
researcher (defined in Chapter 5). One area requiring care in assessment is the
arts (Section 2.6): there is creativity, but the other criteria have to be confirmed
for the activity to qualify as R&D. While routine activity is excluded from R&D,
new methods developed to perform common tasks are included. As an example,
data processing is not an R&D activity unless it is part of a project to develop
new methods for data processing. Vocational training is excluded from R&D, but
new methods to deliver training could be R&D. A new method to fix a problem,
developed as part of a project, could be R&D if the outcome is original and the
other criteria are met.
To be uncertain about the final outcome (uncertain)

2.18 R&D involves uncertainty, which has multiple dimensions. At the
outset of an R&D project, the kind of outcome and the cost (including time
allocation) cannot be precisely determined relative to the goals. In the case of
basic research, which is aimed at extending the boundaries of formal knowledge,
there is a broad recognition of the possibility of not achieving the intended
results. For example, a research project may succeed in eliminating a number
of competing hypotheses, but not all of them. For R&D in general, there is
uncertainty about the costs, or time, needed to achieve the expected results, as
well as about whether its objectives can be achieved to any degree at all. For
example, uncertainty is a key criterion when making a distinction between
R&D prototyping (models used to test technical concepts and technologies with
a high risk of failure, in terms of applicability) and non-R&D prototyping (pre-
production units used to obtain technical or legal certifications).
To be planned and budgeted (systematic)

2.19 R&D is a formal activity that is performed systematically. In this
context “systematic” means that the R&D is conducted in a planned way, with
records kept of both the process followed and the outcome. To verify this, the
purpose of the R&D project and the sources of funding for the R&D performed
should be identified. The availability of such records is consistent with an R&D
project that is aimed at addressing specific needs and has its own human
and financial resources. While the management and reporting structure just
described is more likely to be found in large projects, it can also apply to small-
scale activities where it would be sufficient to have one or more employees or
consultants (providing that a researcher was included) charged with producing a
specific solution to a practical problem.

47
To lead to results that could be possibly reproduced (transferable

and/or reproducible)
2.20 An R&D project should result in the potential for the transfer of the
new knowledge, ensuring its use and allowing other researchers to reproduce
the results as part of their own R&D activities. This includes R&D that has
negative results, in the case that an initial hypothesis fails to be confirmed or
a product cannot be developed as originally intended. As the purpose of R&D is
to increase of the existing stock of knowledge, the results cannot remain tacit
(i.e. remain solely in the minds of the researchers), as they, and the associated
knowledge, would be at risk of being lost. The codification of knowledge and its
dissemination is part of the usual practice in universities and research institutes,
although there may be restrictions for knowledge arising through contract work
or as part of a collaborative undertaking. In a business environment, the results
will be protected by secrecy or other means of intellectual property protection,
but it is expected that the process and the results will be recorded for use by
other researchers in the business.
Examples
2.21 To understand the aim of a project, it is essential to identify its R&D
content and the institutional context in which R&D is performed. Some examples
follow.
● In the field of medicine, a routine autopsy to determine the causes of death
is the practice of medical care and is not R&D; a special investigation of a
particular mortality to establish the side effects of certain cancer treatments
is R&D (in fact, novelty and uncertainty about the final results of the study, as
well as the transferability of the results for broader use, apply here).
● Similarly, routine tests such as blood and bacteriological tests carried out for
medical checks are not R&D, whereas a special programme of blood tests for
patients taking a new drug is R&D.
● Keeping daily records of temperatures or of atmospheric pressure is not R&D,
but a standard procedure. The investigation of new methods of measuring
temperature is R&D, as is the study and development of new models for
weather prediction.
● R&D activities in the mechanical engineering industry often have a close
connection with design. In small and medium-size enterprises (SMEs) in this
industry, there is usually no special R&D department, and R&D performance
is often included under the general heading “design and drawing”. If
calculations, designs, working drawings and operating instructions are
needed for setting up and operating pilot plants or prototypes, they should
be included in R&D. If they are carried out for the preparation, execution
and maintenance of production standardisation (e.g. jigs, machine tools)

or to promote the sale of products (e.g. offers, leaflets, catalogues of spare

parts), they should be excluded from R&D. In this example, several R&D
features can be identified: novelty in exploring the potential of new devices
– by running prototypes; uncertainty, as prototype testing could yield
unexpected results; creativity – emerging in the design of new devices to
be produced; transferability – by producing technical documentation to
translate the results of testing in information to be used at the product
development stage; and a systematic approach, as far as a detailed
organisation of the project can be identified behind the afore-mentioned
technical activities.
2.22 Table 2.1 provides some practical examples of how R&D can be
identified by using the five core criteria.
Table 2.1. Examples of questions for identifying R&D projects

Question Comment
a. What are the objectives The pursuit of original and challenging objectives through the creation of
of the project? “new knowledge” (such as seeking previously undiscovered phenomena,
structures or relationships) is a key criterion for R&D. Any use of already
available knowledge (adaptation, customisation, etc.) which does not entail an
attempt to expand the state of the art should be excluded (Novelty).
b. What is new about this In addition to the development of “new knowledge”, an R&D project should
project? have a creative approach, such as devising new applications of existing
scientific knowledge or new uses of available techniques or technologies
(Creativity).
c. What methods are being Methods used in scientific and technological research, as well as in
used to carry out the research in the social sciences, humanities and the arts, are accepted
project? provided that they address uncertainty about the project’s final outcome.
The uncertainty could be about how much time and resources will be
needed to achieve the planned goal. The choice of method could be part of
the project’s creativity and a means of dealing with uncertainty (Creativity
and uncertainty).
d. How generally applicable To be generally applicable, the findings of an R&D project have to meet
are the findings or results the criterion of being transferable/reproducible, in addition to the other
of the project? four criteria. Transferring the results may for example be demonstrated by
publication in the scientific literature and the use of instruments of intellectual
property protection.
e. What types of staff are A range of skills is assumed to be required to undertake an R&D project
working on the project? (the R&D personnel issue is discussed in Chapter 5 of this manual).
Research personnel in projects are classified as researchers, technicians
and other supporting staff, but only researchers, working as researchers,
are needed to identify an R&D activity which, implicitly, satisfies all five
core criteria.
f. How should the research In selected cases, an “institutional approach” can be used to distinguish
projects of research between R&D and non-R&D projects. For instance, most projects carried
institutions be classified? out in research institutes or research universities can be qualified as R&D
projects. Projects launched in other domains – like business enterprises or
institutions not totally devoted to R&D – should be checked against the five
R&D criteria (see institutions in Chapter 3).

49
2.5. Distribution by type of R&D

2.23 A breakdown by type of R&D is recommended for use in all four of the
sectors used in this manual and defined in Chapter 3: Business enterprise; Higher
education; Government; and Private non-profit. For the purposes of international
comparison, the breakdown could be based either on total R&D expenditure or
on current expenditures only (see Chapter 4). It may be applied at project level,
but some R&D projects may have to be further subdivided.
2.24 There are three types of R&D:
● basic research
● applied research
● experimental development.
Basic research
2.25 Basic research is experimental or theoretical work undertaken
primarily to acquire new knowledge of the underlying foundations of
phenomena and observable facts, without any particular application or use
in view.
2.26 Basic research analyses properties, structures and relationships with
a view to formulating and testing hypotheses, theories or laws. The reference to
no “particular application in view” in the definition of basic research is crucial,
as the performer may not know about potential applications when doing the
research or responding to survey questionnaires. The results of basic research
are not generally sold but are usually published in scientific journals or circulated
to interested colleagues. Occasionally, the publication of basic research may be
restricted for reasons of national security.
2.27 In basic research, the researcher is expected to have some freedom
to set goals. Such research is usually performed in the Higher education sector
but also to some extent in the Government sector. Basic research can be oriented
or directed towards some broad fields of general interest, with the explicit goal
of a range of future applications. Business enterprises in the private sector may
also undertake basic research even though there may be no specific commercial
applications anticipated in the short term. Research on some kinds of energy-
saving technologies may be described as basic according to the above definition
if it does not have a specific use in view. However, it does have a specific direction:
improved energy savings. Such research in this manual is referred to as “oriented
basic research”.
2.28 Oriented basic research may be distinguished from “pure basic
research” as follows:
● Pure basic research is carried out for the advancement of knowledge, without
seeking economic or social benefits or making an active effort to apply the
results to practical problems or to transfer the results to sectors responsible
for their application.

● Oriented basic research is carried out with the expectation that it will produce
a broad base of knowledge likely to form the basis of the solution to recognised
or expected current or future problems or possibilities.
Applied research
2.29 Applied research is original investigation undertaken in order to
acquire new knowledge. It is, however, directed primarily towards a specific,
practical aim or objective.
2.30 Applied research is undertaken either to determine possible uses
for the findings of basic research or to determine new methods or ways of
achieving specific and predetermined objectives. It involves considering the
available knowledge and its extension in order to solve actual problems. In the
Business enterprise sector, the distinction between basic and applied research
is often marked by the creation of a new project to explore promising results of
a basic research programme (moving from a long-term to a medium-short term
perspective in the exploitation of the results of intramural [see Glossary] R&D).
2.31 The results of applied research are intended primarily to be valid
for possible applications to products, operations, methods or systems. Applied
research gives operational form to ideas. The applications of the knowledge
derived can be protected by intellectual property instruments, including
secrecy.
Experimental development
2.32 Experimental development is systematic work, drawing on
knowledge gained from research and practical experience and producing
additional knowledge, which is directed to producing new products or
processes or to improving existing products or processes.
2.33 The development of new products or processes qualifies as
experimental development if it meets the criteria for identifying R&D activity.
An example is uncertainty about the resources needed to achieve the goal of the
R&D project in which the development activity is taking place. In this manual the
“D” in R&D refers to experimental development.
Not “product development”

2.34 The concept of experimental development should not be confused
with “product development”, which is the overall process – from the
formulation of ideas and concepts to commercialisation – aimed at bringing
a new product (good or service) to the market. Experimental development is
just one possible stage in the product development process: that stage when
generic knowledge is actually tested for the specific applications needed to
bring such a process to a successful end. During the experimental development
stage new knowledge is generated, and that stage comes to an end when
the R&D criteria (novel, uncertain, creative, systematic, and transferable

51
and/or reproducible) no longer apply. As an example, in a process aimed at

developing a new car, the option to adopt some technologies could be taken
into consideration and tested for use in the car under development: this is
the stage when experimental development is performed. It will lead to new
results by dealing with new applications of some general knowledge; it will
be uncertain, because testing could give rise to negative results; it will have to
be creative, as the activity will focus on the adaptation of some technology to
a new use; it will be formalised, by needing the commitment of a specialised
workforce; and it will involve a codification, in order to translate the results of
the tests into technical recommendations for the further stages of the product
development process. However, there are cases of product development
without R&D that are discussed in the economics literature, especially in the
case of SMEs.
Not “pre-production development”

2.35 The concept of experimental development should not be confused
with “pre-production development”, which is the term used to describe non-
experimental work on a defence or aerospace product or system before it
goes into production. Similar cases apply in other industries. It is difficult to
define precisely the cut-off point between experimental development and pre-
production development; the distinction between these two categories requires
“engineering judgement” as to when the element of novelty ceases and the work
changes to routine development of an integrated system.
2.36 For example, once a fighter bomber has successfully passed through
the stages of research, technology demonstration, project design and initial
development to the flight-testing of a pre-production aircraft, up to ten additional
airframes may be required in order to ensure full operational integration of the
vehicle into air offence/defence systems. This would be a two-stage process. The
first stage is development of the integrated air offence/defence system, which
involves bringing together developed components and subsystems that have not
previously been integrated in this context. It requires a large flight test programme
for the aircraft, which is potentially very expensive and the main cost element
prior to production. While much of the work commissioned during this stage is
experimental development (R&D), some does not have the element of novelty
necessary for classification as R&D and is instead pre-production development
(non-R&D). The second stage covers trials of the integrated air offence/defence
system. Once the system is proven to work at stage one, the development project
may move on to produce a trial production batch for operational trials (low-rate
initial production). The full production order depends on their success. According
to this manual, this work is not R&D but pre-production development. However,
problems may arise during the trials, and new experimental development may
be needed to solve them. This work is described in this manual as “feedback
R&D” and should be included as R&D.

How types of R&D can be differentiated

2.37 A key criterion guides the classification of R&D activities by type: the
expected use of the results. In addition, two questions can help identify the type
of an R&D project:
● how far ahead in time is the project likely to lead to results that can be applied
● how broad is the range of potential fields of application for the results of the
R&D project (the more fundamental the research, the broader the potential
field of application).
2.38 The relationship between basic research, applied research and
experimental development has to be seen within a dynamic perspective. It is
possible that applied research and experimental development could adapt
fundamental knowledge arising from basic research directly for general
application. However, the linearity of such a process is affected by the feedback
that takes place when knowledge is used to solve a problem. This dynamic
interaction between knowledge generation and the solution of problems links
basic and applied research and experimental development.
2.39 With reference to the organisations where R&D is performed, a
clear-cut separation of the three types of R&D rarely exists. All three types may
sometimes be carried out in the same unit by essentially the same staff, but some
research projects may genuinely straddle categories. For instance, the search for
a new medical treatment for people affected by an epidemic disease may involve
both basic and applied research. It is recommended to undertake an evaluation of
the type of R&D at the project level, by classifying the project’s expected results
according to the two “indicators” described above. Some examples are provided in
the next paragraphs.
Examples of how to differentiate types of R&D in the natural

sciences and engineering
2.40 The following examples illustrate general differences between basic
and applied research and experimental development in the natural sciences and
engineering.
● The study of a given class of polymerisation reactions under various
conditions is basic research. The attempt to optimise one of these reactions
with respect to the production of polymers with given physical or mechanical
properties (making it of particular utility) is applied research. Experimental
development then consists of “scaling up” the process that has been optimised
at the laboratory level and investigating and evaluating possible methods of
producing the polymer as well as products to be made from it.
● The modelling of a crystal’s absorption of electromagnetic radiation is basic
research. The study of the absorption of electromagnetic radiation by this
material under varying conditions (for instance, temperature, impurities,

53
concentration, etc.) to obtain given properties of radiation detection

(sensitivity, rapidity, etc.) is applied research. Testing a new device using this
material in order to obtain a better detector of radiation than those already
existing (in the spectral range considered) is experimental development.
● The development of a new method for the classification of immunoglobulin
sequences is basic research. Investigations undertaken in an effort to
distinguish between antibodies for various diseases is applied research.
Experimental development then consists of devising a method for synthesising
the antibody for a particular disease on the basis of knowledge of its structure
and clinical tests of the effectiveness of the synthesised antibody on patients
who have agreed to accept an experimental advanced treatment.
● A study about how the properties of carbon fibres could change according to
their relative position and orientation within a structure is basic research.
The conceptualisation of a method to allow for processing carbon fibres at
industrial level with a degree of precision at the nano-scale could be the
outcome of some applied research. Testing the use of new composite materials
for different purposes is experimental development.
● Controlling material processes in the domain where quantum effects occur
is an objective to be pursued through basic research. Developing materials
and components for inorganic and organic light-emitting diodes for improved
efficiency and cost reduction is applied research. Experimental development
could be aimed at identifying applications for advanced diodes and
incorporating them in consumer devices.
● Searching for alternative methods of computation, such as quantum
computation and quantum information theory, is basic research. Investigation
into the application of information processing in new fields or in new ways
(e.g. developing a new programming language, new operating systems,
program generators, etc.) and investigation into the application of information
processing to develop tools such as geographical information and expert
systems are applied research. Development of new applications software and
substantial improvements to operating systems and application programmes
are experimental development.
● The study of sources of all kinds (manuscripts, documents, monuments,
works of art, buildings, etc.) in order to better comprehend historical
phenomena (the political, social, cultural development of a country, the
biography of an individual, etc.) is basic research. Comparative analysis of
archaeological sites and/or monuments displaying similarities and other
common characteristics (e.g. geographic, architectural, etc.) to understand
interconnections of potential relevance to teaching material and museum
displays is applied research. The development of new instruments and
methods for studying artefacts and natural objects recovered through
archaeological endeavours (e.g. for the age-dating of bones or botanic
remains) is experimental development.

● In agricultural sciences and forestry:
❖ Basic research: Researchers investigate genome changes and mutagenic

factors in plants to understand their effects on the phenome. Researchers
investigate the genetics of the species of plants in a forest in an attempt
to understand natural controls for disease or pest resistance.
❖ Applied research: Researchers investigate wild potato genomes to locate
the genes responsible for resistance to potato blight in an effort to
improve the disease resistance in domestic/crop potatoes. Researchers
plant experimental forests where they alter the spacing and alignment
of the trees to reduce the spread of disease while ensuring the optimum
arrangement for maximum yield.
❖ Experimental development: Researchers create a tool for gene editing
by using knowledge of how enzymes edit DNA. Researchers use existing
research on a specific plant species to create a plan for improving how a
company plants its forests to achieve a specific goal.
● In nanotechnology:
❖ Basic research: Researchers study the electrical properties of graphene

by using a scanning tunnelling microscope to investigate how electrons
move in the material in response to voltage changes.
❖ Applied research: Researchers study microwaves and thermal coupling
with nanoparticles to properly align and sort carbon nanotubes.
❖ Experimental development: Researchers use research in micro-
manufacturing to develop a portable and modular micro-factory system
with components that are each a key part of an assembly line.
● In computer and information sciences:
❖ Basic research: Research on the properties of general algorithms for

handling large amounts of real-time data.
❖ Applied research: Research to find ways to reduce the amount of spam
by understanding the whole structure or business model of spam, what
spammers do, and their motivations in spamming.
❖ Experimental development: A start-up company takes code developed
by researchers and develops the business case for the resulting software
product for improved on-line marketing.
Examples of how to differentiate types of R&D in the social

sciences, humanities and the arts
2.41 Another set of examples can be provided with reference to the social
sciences, humanities and the arts where, as discussed above, the blurring of
boundaries could affect the distinction between basic and applied research.
Examples of experimental development in these domains can also be difficult to
identify, because of the role played by other domains in the natural sciences and

55
engineering. It should be noted that these examples must also meet the basic
criteria identified in this chapter to be considered as R&D.
● In economics and business:
❖ Basic research: A review of theories on the factors determining regional

disparities in economic growth. Economists conducting abstract research
in economic theory that focuses on whether a natural equilibrium exists
in a market economy. The development of new risk theories.
❖ Applied research: The analysis of a specific regional case for the purpose
of developing government policies. Economists investigating the
properties of an auction mechanism that could be relevant to auctioning
the telecommunications spectrum. The investigation of new types of
insurance contracts to cover new market risks or new types of savings
instruments.
❖ Experimental development: The development of operational models,
based upon statistical evidence, to design economic policy tools to
allow a region to catch up in terms of growth. The development by a
national telecommunications authority of a method for auctioning the
telecommunications spectrum. The development of a new method to
manage an investment fund is experimental development as long as
there is sufficient evidence of novelty.
● In education:
❖ Basic research: Analysis of the environmental determinants of learning

ability. The investigation by researchers of the effect of different types of
manipulatives on the way first graders learn mathematical strategy by
changing manipulatives and then measuring what students have learned
through standardised instruments
❖ Applied research: The comparative evaluation of national education
programmes aimed at reducing the learning gap experienced
by disadvantaged communities. The study by researchers of the
implementation of a specific math curriculum to determine what
teachers needed to know to implement the curriculum successfully.
❖ Experimental development: The development of tests for selecting which
educational programme should be used for children with specific needs.
The development and testing (in a classroom) of software and support
tools, based on fieldwork, to improve mathematics cognition for student
special education.
● In social and economic geography:
❖ Basic research: Researchers seek to understand the fundamental

dynamics of spatial interactions.
❖ Applied research: A research study analyses the spatial-temporal patterns
in the transmission and diffusion of an infectious disease outbreak.

● In history:
❖ Basic research: Historians study the history and human impact of glacial
outburst floods in a country.
❖ Applied research: Historians examine past societies’ responses to
catastrophic natural events (e.g. floods, droughts, epidemics) in order to
understand how contemporary society might better respond to global
climate change.
❖ Experimental development: Using previous research findings, historians
design a new museum exhibit on the adaptations of past human societies
to environmental changes; this serves as a prototype for other museums
and educational installations.
● In language/linguistics:
❖ Basic research: Linguists study how different languages interact as they

come into contact with one another.
❖ Applied research: Speech therapists examine the governing neurology of
languages and how humans acquire language skills.
❖ Experimental development: Linguists develop a tool for diagnosing
autism in children based on their language acquisition, retention and use
of signs.
● In music:
❖ Basic research: Researchers develop a transformational theory that

provides a framework for understanding musical events not as a
collection of objects that have particular relationships to each other but
as a series of transformational operations applied to the basic material
of the work.
❖ Applied research: Researchers use historical records and the techniques
of experimental archaeology to recreate an ancient and long-disappeared
musical instrument and to determine how it would have been constructed,
how it was played and the types of sounds it would have produced.
❖ Experimental development: Music educators and theorists work
to produce new pedagogical materials based on new discoveries in
neuroscience that change our understanding of how humans process
new sounds and information.
2.6. Classification and distribution by Fields of Research

and Development (FORD)
2.42 For a number of reasons, survey practitioners and data users often
find it helpful and relevant to classify R&D-performing units and distribute
their R&D resources according to the knowledge domain in which they operate.
This manual proposes the use of the OECD Fields of Research and Development
(FORD) classification for such purposes. This classification, developed for R&D

57
measurement purposes, follows primarily a content approach. Where the content

of the R&D subject matter is closely related, subjects are grouped together to form
the broad (one-digit) and narrower (two-digit) fields of the classification. While
the classification can be applied to a broader range of science and technology
(S&T) and knowledge-based activities, its formulation by the OECD is focused on
R&D as defined in this manual.
2.43 The aim is to distribute R&D efforts and classify the units that
undertake such efforts. Two R&D projects can be said to belong to the same field
if their content is the same or sufficiently similar. The following criteria give rise
to the FORD classification and can help inform the assessment of the degree of
similarity of the subject matter content:
● The knowledge sources drawn upon for the R&D activity carried out. The
application of developments in some technology fields often gives rise to new
scientific efforts, in the same way that scientific knowledge provides a basis
for new technological developments.
● The objects of interest – the phenomena to be understood or the problems to
be solved as part of R&D.
● The methods, techniques and professional profiles of the scientists and other
R&D workers – different domains can be distinguished sometimes on the
basis of the methodological approaches to the study of a given phenomenon
or question.
● The areas of application. For example, in the FORD classification, the medical
sciences and agricultural sciences are specifically defined by their applications
to human health and agricultural activities.
2.44 This classification is closely related to and consistent with UNESCO’s
“Recommendation concerning the International Standardisation of Statistics
on Science and Technology” (UNESCO, 1978), which provided the initial basis for
the OECD classification of R&D by the field of S&T in previous versions of this
manual. There is also a degree of relationship with the ISCED Fields of Education
and Training (ISCED-F), which is aimed at the classification of study and training
programmes and reflects to a large extent the way in which schools, departments,
etc., organise their activities and award credentials to students who successfully
complete these programmes. It is recognised that FORD and the ISCED-F have
different purposes, and it is not feasible to ensure a direct correspondence
between the two classifications (UNESCO-UIS, 2014, p. 17).
2.45 In light of ongoing changes in the way R&D is conducted and the
progressive emergence of new domains, the FORD classification will be subject
to continuing revision after the publication of this manual’s edition. For more up-
to-date versions, the reader should consult the online annexes to this manual
where more detail can be found on this classification and its use.

Table 2.2. Fields of R&D classification

Broad classification Second-level classification
1. Natural sciences 1.1 Mathematics

1.2 Computer and information sciences
1.3 Physical sciences
1.4 Chemical sciences
1.5 Earth and related environmental sciences
1.6 Biological sciences
1.7 Other natural sciences
2. Engineering and technology 2.1 Civil engineering
2.2 Electrical engineering, electronic engineering,
information engineering
2.3 Mechanical engineering
2.4 Chemical engineering
2.5 Materials engineering
2.6 Medical engineering
2.7 Environmental engineering
2.8 Environmental biotechnology
2.9 Industrial biotechnology
2.10 Nano-technology
2.11 Other engineering and technologies
3. Medical and health sciences 3.1 Basic medicine
3.2 Clinical medicine
3.3 Health sciences
3.4 Medical biotechnology
3.5 Other medical science
4. Agricultural and veterinary sciences 4.1 Agriculture, forestry, and fisheries
4.2 Animal and dairy science
4.3 Veterinary science
4.4 Agricultural biotechnology
4.5 Other agricultural sciences
5. Social sciences 5.1 Psychology and cognitive sciences
5.2 Economics and business
5.3 Education
5.4 Sociology
5.5 Law
5.6 Political science
5.7 Social and economic geography
5.8 Media and communications
5.9 Other social sciences
6. Humanities and the arts 6.1 History and archaeology
6.2 Languages and literature
6.3 Philosophy, ethics and religion
6.4 Arts (arts, history of arts, performing arts, music)
6.5 Other humanities

59
2.7. Examples of R&D, boundaries and exclusions

in different areas
R&D and innovation activities and borderline cases
2.46 Innovation is currently defined for measurement purposes in the
third edition of the Oslo Manual (OECD/Eurostat, 2005) with a sole focus on
the Business enterprise sector (see a definition for this sector in Chapter 3).
In summary, it has to do with putting new or significantly improved products
on the market or finding better ways (through new or significantly improved
processes and methods) of getting products to the market. R&D may or may not
be part of the activity of innovation, but it is one among a number of innovation
activities. These activities also include the acquisition of existing knowledge,
machinery, equipment and other capital goods, training, marketing, design and
software development. These innovation activities may be carried out in-house
or procured from third parties.
2.47 Care must be taken to exclude activities that, although part of the
innovation process, do not satisfy the criteria required to be classified as
R&D. For example, patent application and licensing activity, market research,
manufacturing start-up, and tooling up and redesign for the manufacturing
process are not in their own right R&D activities and cannot be assumed
to be part of an R&D project. Some activities, such as tooling up, process
development, design and prototype construction, may contain an appreciable
element of R&D, making it difficult to identify precisely what should or should
not be defined as R&D. This is particularly true for defence and large-scale
industries such as aerospace. Similar difficulties may arise in distinguishing
public technology-based services such as the inspection and control of food
and drugs from related R&D.
Examples of what is or is not R&D in innovation processes

2.48 Table 2.3, and the examples that follow, provide more information on
what should and what should not be counted as R&D.
Prototypes
2.49 A prototype is an original model constructed to include all the
technical and performance characteristics of the new product. For example, if
a pump for corrosive liquids is being developed, several prototypes are needed
for accelerated life tests with different chemicals. A feedback loop exists so
that if the prototype tests are not successful, the results can be used for further
development of the pump.
2.50 The design, construction and testing of prototypes normally falls

within the scope of R&D. This applies whether only one or several prototypes
are made and whether they are made consecutively or simultaneously. However,
when any necessary modifications to the prototype(s) have been made and
testing has been satisfactorily completed, the end-point of R&D has been

reached. The construction of several copies of a prototype to meet a temporary

commercial, military or medical need after successful testing of the original, even
if undertaken by R&D staff, is not part of R&D. The virtualisation of prototyping
could follow the same rules, and it will be included in R&D as far as the testing
activity is part of an R&D project and is aimed at collecting evidence essential for
achieving the objectives of the project.
Table 2.3. Borderline between R&D, innovation

and other business activities
Item Treatment Remarks
Prototypes Include in R&D As long as the primary objective is to make

further improvements.
Pilot plant Include in R&D As long as the primary purpose is R&D.
Industrial design Split Include design required during R&D. Exclude
design for production process.
Industrial engineering and Split Include “feedback” R&D and tooling up

tooling up industrial engineering in innovation processes.
Exclude for production processes.
Trial production Split Include if production implies full-scale

testing and subsequent further design and
engineering. Exclude all other associated
activities.
Pre-production development Exclude
After-sales service and trouble- Exclude Except “feedback” R&D (to be included).
shooting
Patent and licence work Exclude All administrative and legal work needed to
apply for patents and licences (delivering
documentation as an outcome of R&D
projects is R&D). However, patent work
connected directly with R&D projects is R&D.
Routine tests Exclude Even if undertaken by R&D personnel.
Data collection Exclude Except when an integral part of R&D.
Routine compliance with public Exclude
inspection control, enforcement
of standards, regulations
Pilot plants
2.51 The construction and operation of a pilot plant is a part of R&D as long
as the principal purposes are to obtain experience and to compile engineering
and other data to be used in:
● evaluating hypotheses
● writing new product formulas
● establishing new finished product specifications
● designing special equipment and structures required by a new process
● preparing operating instructions or manuals on the process.

61
2.52 If, as soon as this experimental phase is over, a pilot plant switches to
operating as a normal commercial production unit, the activity can no longer be
considered R&D even though it may still be described as a pilot plant. As long as
the primary purpose in operating a pilot plant is non-commercial, it makes no
difference in principle if part or all of the output is sold. Such receipts should not
be deducted from the cost of R&D activity (Chapter 4).
Large-scale projects
2.53 Large-scale projects (in areas like defence, aerospace or big science)
usually cover a spectrum of activity from experimental to pre-production
development. Under these circumstances, the funding and/or performing
organisation often cannot distinguish between R&D and other elements
of expenditure. The distinction between R&D and non-R&D expenditure is
particularly important in countries where a large proportion of government R&D
expenditure is directed to defence.
2.54 It is important to look closely at the nature of costly pilot plants
or prototypes, such as the first of a new line of nuclear power stations or of
icebreakers. They may be constructed almost entirely from existing materials
and use existing technology, and they are often built for simultaneous use for
R&D and for providing the primary service concerned (power generation, ice
breaking). The construction of such plants and prototypes should not be wholly
credited to R&D. Only the additional costs due to the experimental nature of
these products should be attributed to R&D.
Trial production
2.55 After a prototype has been satisfactorily tested and any necessary
modifications made, the manufacturing start-up phase may begin. This
is related to full-scale production; it may consist of product or process
modification or retraining personnel in the new techniques or in the use of new
machinery. Unless the manufacturing start-up phase implies further design
and engineering R&D, it should not be counted as R&D, since the primary
objective is no longer to make further improvements to the products but to
start the production process. The first units of a trial production run for a mass
production series should not be considered as R&D prototypes even if they are
loosely described as such.
2.56 For example, if a new product is to be assembled by automatic
welding, the process of optimising the settings on the welding equipment in
order to achieve maximum production speed and efficiency would not count
as R&D.
Trouble-shooting
2.57 Trouble-shooting occasionally shows the need for further R&D, but
more frequently it involves the detection of faults in equipment or processes and
results in minor modifications of standard equipment and processes. It should
not, therefore, be included in R&D.

“Feedback” R&D
2.58 After a new product or process has been turned over to production
units, there will still be technical problems to be solved, some of which may
demand further R&D. Such “feedback” R&D should be included.
Tooling up and industrial engineering
2.59 In most cases, the tooling up and industrial engineering phases of
any project are considered to be part of the production process, and not of R&D.
Three phases of tooling up can be identified:
● the first-time use of components (including the use of components resulting
from R&D efforts)
● the initial tooling of equipment for mass production
● installing equipment linked with the start of mass production.
2.60 If the tooling up process results in further R&D work, such as

improvements in the production of machinery and tools or changes to the
production and quality control procedures or the development of new methods
and standards, these activities are classified as R&D. “Feedback” R&D resulting
from the tooling up phase should be defined as R&D.
Clinical trials
2.61 Before new drugs, vaccines, devices or treatments can be introduced
onto the market, they must be tested systematically on human volunteers to
ensure that they are both safe and effective. These clinical trials are divided into
four standard phases, three of which take place before permission to manufacture
is granted. For the purposes of international comparison, by convention, clinical
trial phases 1, 2 and 3 can be treated as R&D. Phase 4 clinical trials, which continue
testing the drug or treatment after approval and manufacture, should only be
treated as R&D if they bring about a further scientific or technological advance.
Moreover, not all activities undertaken prior to permission to manufacture
are considered to be R&D, especially when there is a significant wait after the
completion of phase 3 trials, during which activities related to marketing and
process development may be started.
R&D and design

2.62 Design and R&D activities are difficult to separate. Some design
activities are an integral part of R&D projects, and R&D can be an input to
new design efforts. There are similarities and linkages. However, not all
design meets the functional novelty and uncertainty tests as captured in this
chapter’s five core R&D criteria. Design plays a key role in the development
and implementation of innovations. As an agreed definition of design for
statistical purposes does not yet exist, design can be described as a potential
multi-faceted innovation activity aimed at planning and designing procedures,
technical specifications and other user and functional characteristics for

63
new products and processes. Among these activities are initial preparations
for the planning of new products or processes, and work on their design and
implementation, including adjustments and further changes. This description
emphasises the creative role of design within an innovation process, a feature
potentially shared with the R&D performed in the same context. Some design-
related activities may be considered R&D to the extent that they play a role
in a product development process, which is aiming at something “new” (but
not necessarily at new knowledge), is creative and original, can be formalised
(performed by a dedicated team), and leads to a codified output to be passed on
to the development team. The main difference with R&D is that no uncertainty
is likely to be found when skilled designers are asked to contribute to an
innovation project. This leads to a view that design is not R&D and that it has to
be kept distinct from R&D for any statistical purpose.
2.63 While an R&D project involves uncertainty about whether an
expected outcome will be delivered within an agreed time schedule, a design
project’s uncertainty will be directly influenced by the clarity and the feasibility
of its original goals. As an example, designing a standard building does not
involve major uncertainty about the final outcome; yet the more challenging
the concept of the building, the adding of new features, for example, the higher
the uncertainty about the time and costs needed to complete the project. R&D
activity, complementing the use of existing design tools, may be required to
address the uncertainty.
R&D and artistic creation

2.64 Design sometimes tends to be characterised by the use of artistic
methods. This is another potential area of overlap. In order to address the
discussion of R&D and artistic creation, it can be useful to make a distinction
between research for the arts, research on the arts and artistic expression.
Research for the arts

2.65 Research for the arts consists in developing goods and services to
meet the expressive needs of artists and performers. There are enterprises in
this line of business that devote a significant part of their resources to R&D in
this area. For instance, they engage in experimental development to produce new
electronic musical instruments to suit the needs of a group of performers. Other
types of R&D organisations (mainly universities and technical institutes) also
play a role in exploring new technologies for performance art (to improve audio/
video quality, for instance). The activity aimed at supporting the introduction
of new organisational or marketing methods by art institutions (advertising,
financial management, etc.) may qualify as R&D, but caution should be exercised
in making this decision. This area of R&D performance is already covered by
existing data collection.

Research on the arts (studies about the artistic expression)

2.66 Basic or applied research contributes to most of the studies of the arts
(musicology, art history, theatre studies, media studies, literature, etc.). Public
research institutions could have a role in selected research domains (as some
relevant research infrastructures – like libraries, archives, etc. – are often attached
to arts institutions, such as museums, theatres, etc.). As far as preservation and
restoration activities are concerned (if not to be included in the group above),
it is recommended to identify the providers of such technical services as R&D
performers (employing researchers, publishing scientific works, etc.). This area
of R&D performance is largely covered by existing data collection.
Artistic expression versus research

2.67 Artistic performance is normally excluded from R&D. Artistic
performances fail the novelty test of R&D as they are looking for a new
expression, rather than for new knowledge. Also, the reproducibility criterion
(how to transfer the additional knowledge potentially produced) is not met. As a
consequence, arts colleges and university arts departments cannot be assumed
to perform R&D without additional supporting evidence. The existence of artists
attending courses in such institutions is not relevant to the R&D measurement.
Higher education institutions have, nevertheless, to be evaluated on a case-
by-case basis if they grant a doctoral degree to an artist as a result of artistic
performances. The recommendation is to adopt an “institutional” approach and
only to take account of artistic practice recognised as R&D by higher education
institutions as potential R&D (to be further used by data collectors).
R&D and software development

2.68 Information technology has a pervasive role in almost every innovation
activity, and largely relies on R&D activities but also influences the ability of
enterprises and institutions to perform R&D effectively. Software development
is an innovation-related activity that is sometimes connected with R&D and
incorporates, under specific conditions, some R&D. For a software development
project to be classified as R&D, its completion must be dependent on a scientific
and/or technological advance, and the aim of the project must be the systematic
resolution of a scientific and/or technological uncertainty.
2.69 In addition to the software that is part of an overall R&D project (to
record and monitor its different stages, for instance), the R&D associated with
software as an end product or software embedded in an end product could also
be classified as R&D when the R&D criteria apply.
2.70 The nature of software development is such that it is difficult to
identify its R&D component, if any. Software development is an integral part
of many projects that in themselves have no element of R&D. The software
development component of such projects, however, may be classified as R&D
if it leads to an advance in the area of computer software. Such advances are

65
generally incremental rather than revolutionary. Therefore, an upgrade, addition

or change to an existing program or system may be classified as R&D if it
embodies scientific and/or technological advances that result in an increase in
the stock of knowledge. The use of software for a new application or purpose
does not by itself constitute an advance.
2.71 The following examples illustrate the concept of R&D in software and
should be included in R&D:
● the development of new operating systems or languages
● the design and implementation of new search engines based on original

technologies
● the effort to resolve conflicts within hardware or software based on the process
of re-engineering a system or a network
● the creation of new or more efficient algorithms based on new techniques
● the creation of new and original encryption or security techniques.
2.72 Software-related activities of a routine nature are not to be

considered R&D. Such activities include work on system-specific or program-
specific advances that were publicly available prior to the commencement of the
work. Technical problems that have been overcome in previous projects on the
same operating systems and computer architecture are also excluded. Routine
computer and software maintenance are not included in R&D.
Examples of other software-related activities to be excluded
from R&D are:
● the development of business application software and information systems
using known methods and existing software tools
● adding user functionality to existing application programs (including basic
data entry functionalities)
● the creation of websites or software using existing tools
● the use of standard methods of encryption, security verification and data

integrity testing
● the customisation of a product for a particular use, unless during this process
knowledge is added that significantly improves the base program
● routine debugging of existing systems and programs, unless this is done prior
to the end of the experimental development process.
2.73 In the systems software area, individual projects may not be
considered as R&D, but their aggregation into a larger project could generate some
technological uncertainty, the resolution of which will need R&D. Alternatively,
a large project can be aimed at developing a commercial product by adopting
available technologies and not include R&D in its planning, but there may be
some elements in the project that would need some additional R&D activity to
assure the smooth integration of different technologies.

2.74 Starting in the 1993 SNA (EC et al., 1994), total expenditure on software
(including R&D for software development) was regarded as capital investment.
In the 2008 SNA (EC et al., 2009), total expenditure on R&D is regarded as capital
investment. According to the Handbook on deriving capital measures of intellectual
property products (OECD, 2009), which further developed the 2008 SNA guidance
on intangibles, capitalised software R&D remained in software investment. It is
important to be able to identify explicitly R&D expenditure devoted to software to
better inform both R&D and SNA statisticians and users of the overlap between
software and R&D. This is discussed further in Chapter 4.
R&D and education and training

2.75 Educational and training institutions below the tertiary level focus their
resources on teaching and, as a result, have a very low likelihood of being involved
in R&D projects. On the other hand, in higher education institutions research and
teaching are always very closely linked, as most academic staff undertake both,
and many buildings, as well as much equipment, serve both purposes.
2.76 Because the results of research feed into teaching, and because the
information and experience gained in teaching can often result in an input to
research, it is difficult to define where the education and training activities of
higher education staff and their students end and R&D activities begin, and
vice versa. R&D’s elements of novelty distinguish it from routine teaching
and other work-related activities. The adoption of the key R&D criteria can be
supplemented, in this sector, by a consideration of the institutional role played
by some actors:
● doctoral students and master’s students meeting specific conditions (Chapters 5
and 9)
● supervisors of students (included in the university staff)
● providers of specialised health care in university hospitals.
2.77 Since the research activity performed by doctoral students should

be included in the overall R&D performed by the Higher Education sector, both
they and the university staff acting as their instructors or supervisors should be
included in R&D personnel totals (see Chapter 5). Of course, the time spent by
the university staff to undertake tasks that are not related to research should be
excluded from the estimation of the actual R&D performance. This applies to all
scientific disciplines.
2.78 Similarly, in university hospitals where the training of medical
students is an important activity in addition to the primary activity of health
care, the activities of teaching, R&D and advanced, as well as routine, medical
care are frequently closely linked. Personnel and students providing specialised
health care in university hospitals, when they can be safely assumed to be part of
an overall R&D effort, should be included in R&D personnel. Any routine activity
undertaken to provide health care in the same context should be excluded
from R&D.

67
R&D in service activities

2.79 The 2008 SNA defines services as the result of a production activity
that changes the conditions of the consuming units, or facilitates the exchange
of products or financial assets. Among the former, service providers can
affect changes in the condition of the consumer’s goods, in the physical and
mental conditions of persons (e.g. through health or transportation, as well as
through the provision of information, education, etc.). The SNA also defines a
separate hybrid product category that has features of both goods and services,
namely “Knowledge-capturing products”. These concern the provision, storage,
communication and dissemination of information, advice and entertainment
in such a way that the consuming unit can access the knowledge repeatedly.
The industries that produce these products are those concerned with the
provision, storage, communication and dissemination of information, advice
and entertainment in the broadest sense of those terms (EC et al., 2009).
2.80 The provision of services entails a high degree of proximity and
interaction with customers. In addition, industries that specialise in the
production of goods may be actively involved in the delivery of services. Firms in
the service industries may in turn control several aspects of goods production,
including for example the experimental development of new goods that are part
of their service delivery.
2.81 Defining the boundaries of R&D in service activities is therefore
difficult, for two main reasons: first, it is difficult to identify projects involving
R&D that is specific to a service and not embedded in a good or knowledge-
capturing product; and, second, the line between R&D and other innovation
activities is not always clear.
2.82 Among the many projects in services, those that constitute R&D result
in new knowledge or the use of knowledge to devise new applications, in keeping
with the definition of R&D.
2.83 Identifying R&D is more difficult in service activities than in goods-
producing industries because the R&D is not necessarily specialised in a field of
research, although there may be specialisation reflecting the market served. R&D
covers several areas: technology-related R&D, and R&D in the social sciences,
humanities and arts, including R&D relating to the knowledge of behaviour and
organisations. This last notion is already included in the criterion “knowledge of
humankind, culture and society”, but it is particularly important in the case of
service activities. Because these forms of R&D may be combined in a project, it
is important to define clearly the various forms of R&D involved. If the analysis
is confined to technology-related R&D, for example, R&D may be understated. In
many cases, R&D findings in service industries may be part of service delivery
activities.

2.84 Also, in service companies, R&D is not always organised as formally as

in goods-producing companies (i.e. with a dedicated R&D department, researchers
or research engineers identified as such in the establishment’s personnel list,
etc.). The concept of R&D in services is still less specific and sometimes goes
unrecognised by the enterprises involved. As more experience becomes available
on the collection of R&D data in services, the criteria for identifying R&D and
examples of service-related R&D may require further development.
Criteria for identifying R&D in services

2.85 In addition to the five core criteria, the following are indicators that
may help to identify the presence of R&D in service activities:
● links with public research laboratories
● the involvement of staff with doctoral degrees or doctoral students
● the publication of research findings in scientific journals, the organisation of

scientific conferences or involvement in scientific reviews.
Examples of R&D in selected service activities
2.86 The R&D activities listed below may serve as examples of R&D in
service activities. The general and supplementary criteria for distinguishing R&D
presented in Section 2.4 also have to be taken into account.
2.87 The general boundaries of R&D as defined above largely apply to
service activities. The element of novelty is a basic criterion for distinguishing
R&D from related activities.
Examples of R&D in banking and insurance
● mathematical research relating to financial risk analysis
● the development of risk models for credit policy
● the experimental development of new software for home banking
● the development of techniques for investigating consumer behaviour for the

purpose of creating new types of accounts and banking services
● research to identity new risks or new characteristics of risk that need to be
taken into consideration in insurance contracts
● research on social phenomena with an impact on new types of insurance
(health, retirement, etc.), such as on insurance cover for non-smokers
● R&D related to electronic banking and insurance, Internet-related services
and e-commerce applications
● R&D related to new or significantly improved financial services (new concepts
for accounts, loans, insurance and saving instruments).

69
Examples of R&D in some other service activities

● analysis of the effects of economic and social change on consumption and
leisure activities
● the development of new methods for measuring consumer expectations and
preferences
● the development of new methods to deliver and measure social service
outcomes that can be adapted in a variety of different socioeconomic or
cultural settings
● the development of new survey methods and instruments
● the development of tracking and tracing procedures (logistics)
● research into new travel and holiday concepts.
R&D and related scientific and technological activities

2.88 Difficulties in separating R&D from other scientific and technological
activities (STA) arise when several activities are performed in the same institution.
In data collection practices, criteria are usually applied on the basis of a direct
knowledge of the performing institutions. As general guidelines:
● Institutions or units of institutions and firms whose principal activity is
R&D often have secondary, non-R&D activities (e.g. scientific and technical
information, testing, quality control, analysis). Insofar as a secondary activity
is undertaken primarily in the interests of R&D, it should be included in R&D;
if the secondary activity is designed essentially to meet needs other than R&D,
it should be excluded.
● Institutions whose main purpose is an R&D-related scientific activity often
undertake some research in connection with this activity. Such research
should be isolated and included when measuring R&D.
2.89 In some sectors, the key criteria for distinguishing between R&D
and related scientific and technological activities are particularly difficult to
apply. General-purpose data collection, testing and standardisation, big data
projects, space exploration, and mineral exploration and evaluation are all areas
involving large amounts of resources, and any variations in their treatment will
have important effects on the international comparability of the resulting R&D
data. Large-scale projects also pose problems for the identification of their R&D.
As this edition is being finalised, UNESCO is updating its definitions of STA for
statistical purposes (UNESCO, 1978; UNESCO, 1984), and the outcome of that
process is expected to provide further guidance on the boundaries between R&D
and other STA activities. Such guidance will be made available in due course as
part of the online annex material to this manual.
General-purpose data collection and documentation

2.90 General-purpose data collection is undertaken generally by government
agencies to record natural, biological or social phenomena that are of general public

interest or that only the government has the resources to record. Examples are
routine topographical mapping; routine geological, hydrological, oceanographic
and meteorological surveying; and astronomical observations. Data collected
solely or primarily as part of the R&D process are included in R&D (e.g. data
collected by a detector that is part of an elementary particle scattering experiment
at CERN). The same reasoning applies to the processing and interpretation of the
data. The social sciences, in particular, are very dependent on an accurate record of
facts relating to society in the form of censuses, sample surveys, etc. When these
are specially collected or processed for the purpose of scientific research, the cost
should be attributed to research and should cover the planning and systematising
of the data. R&D can also be identified when a specific project is aimed at
developing totally new statistical methods (e.g. conceptual and methodological
work in relation to the development of completely new or substantially modified
surveys and statistical systems, work on sampling methodologies, small area
statistical estimates and advanced data-capturing techniques) or data collection
methodologies and techniques. However, data collected for other or general
purposes, such as quarterly sampling of unemployment, should be excluded from
R&D even if exploited for research (unless the researcher had to pay for the right
to use such data in the research). Market surveys should also be excluded.
2.91 The activities of a scientific and technical information service or of a
research laboratory library that is maintained predominantly for the benefit of
the research workers in the laboratory should be included in R&D. The activities
of a firm’s documentation centre open to all the firm’s staff should be excluded
from R&D even if it shares the same premises as the business research unit
(the need to avoid an over-evaluation of R&D-related activities applies here).
Similarly, the activities of central university libraries should be excluded from
R&D. These criteria, which will have to be applied also to electronic libraries and
data repositories, apply only when it is necessary to deal with the activities of
an institution or a department in their entirety. Where more detailed accounting
methods are used, it may be possible to impute part of the costs of the excluded
activities to R&D overheads. Whereas the preparation of scientific and technical
publications is, generally speaking, excluded, the preparation of the original
report of research findings should be included in R&D.
Testing and standardisation

2.92 Public bodies and consumer organisations often operate laboratories
that are intended mainly to test products and verify that standards are met. In
addition to standard testing and benchmarking activities – which are not R&D –
the staff of these laboratories may also spend time devising new or substantially
improved testing methods. Such activities should be included in R&D.
Big data projects

2.93 The advent of new instruments and methods of data-intensive
exploration is facilitating the process of data-intensive scientific discovery

71
and data-driven innovation. These activities are R&D if and only if they meet
the five core criteria, in particular the general requirement that the activity or
project should be undertaken in a systematic way – i.e. by clearly identifying
the original knowledge gap and focusing specific resources on addressing it.
An example is the “Human Genome Project”, which attracted researchers and
institutions from 18 countries to co-operate in a 13-year-long research effort
to sequence and map out the human DNA code. Through digitisation, the R&D
codification criterion plays a major part in big data projects, as the usability of
the data arising from “big data” science projects depends on its ability to convey
knowledge about specific phenomena for which the data have been gathered.
These data may or may not be made widely accessible or usable for research
purposes. The concept of open science commonly refers to efforts to make the
output of publicly funded research more widely accessible in digital format to
the scientific community, the business sector or society more generally (OECD,
2015). In some cases, efforts to make research data openly accessible to the
broad scientific community, including developing specific tools that facilitate
the reproducibility of the research, will be an integral part of an R&D project,
provided that they are explicitly formulated as such within the R&D project’s
objectives and are budgeted. In other cases, these should be treated as separate
dissemination efforts and not counted as R&D.
Space exploration
2.94 A difficulty with space exploration is that, in some respects, much
space activity may now be considered routine; certainly, most costs are incurred
for the purchase of goods and services that are not R&D. It may be necessary
to separate the activities associated with space exploration, including the
development of vehicles, equipment, software and techniques, from those
involved in the routine placing of orbiting satellites or the establishment of
tracking and communication stations.
Mineral exploration and evaluation

2.95 Mineral exploration and evaluation is defined in the 2008 SNA as
a category of activity leading to the creation of an intellectual property asset,
separate from R&D (EC et al., 2009; OECD, 2009). The activity of mineral exploration
and evaluation adds to the knowledge of subsoil deposits in specific locations
for the purpose of their economic exploitation. It includes the acquisition
of exploration rights as well as topographical, geological, geochemical and
geophysical studies, and trenching, sampling and evaluation activities.
2.96 This manual also makes a strict separation between R&D and mineral
exploration. However, some links with R&D can be found. For example, a number
of geological tests undertaken in the context of R&D projects can provide initial
evidence for exploration and mining companies to follow up with exploitation-
driven exploration efforts, which are not R&D. R&D may also be required to

develop new test and drilling techniques that the mining industry can draw on
for its exploration and routine activities. Mining and prospecting sometimes
cause problems owing to a linguistic confusion between research for new or
substantially improved resources (food, energy, etc.) and the search for existing
reserves of natural resources, a confusion that blurs the distinction between
R&D and surveying and prospecting. In theory, in order to establish accurate R&D
data, the following activities should be identified, measured and summed:
● the development of new surveying methods and techniques
● surveying undertaken as an integral part of a research project on geological

phenomena
● research on geological phenomena, undertaken as a subsidiary part of
surveying and prospecting programmes.
2.97 In practice, the last of these presents a number of problems. It is
difficult to frame a precise definition that would be meaningful to respondents to
national surveys. For this reason, only the following activities should be included
in R&D:
● the development of new or substantially improved methods and equipment
for data acquisition and for the processing and study of the data collected and
for the interpretation of these data
● surveying undertaken as an integral part of an R&D project on geological
phenomena, including data acquisition, processing and interpretation
undertaken for primarily scientific purposes.
2.98 It follows that the surveying and prospecting activities of commercial
companies will be almost entirely excluded from R&D. For example, the sinking
of exploratory wells to evaluate technological services is not R&D.
R&D and the management of science and technology (S&T) activities

Technology readiness levels
2.99 The classification of large R&D projects is discussed in Chapter 8, with
an emphasis on the defence and aerospace industries. In some jurisdictions,
classifications of the technology readiness level (TRL) are used in procurement
and in the description of projects. As there are a number of such classifications,
the recommendation is that, if there is one in use in the jurisdiction of interest, it
can be assessed to determine whether it could make a contribution to improving
the collection of R&D statistics.
Demonstration projects
2.100 Two concepts of demonstration have already been adopted in R&D
statistics: “user demonstration”, which takes place when a prototype is operated
at or near full scale in a realistic environment to aid the formulation of policy
or the promotion of its use, which is not R&D; and “technical demonstration”

73
(including the development of “demonstration projects” and “demonstration

models”) which, because it is an integral part of an R&D project, is an R&D activity.
2.101 With reference to its broad use in the management of large
research projects, “technology demonstration” is seen as a step in the process
of evaluating, ex-ante or ex-post, the implementation of new technologies. This
meaning was originally adopted in the information and communication sector
and has evolved to mean the activity carried out to show to potential investors
and customers the expected potentiality of a technology under development. In
this respect, the use of this concept is not recommended in association with the
R&D concept, unless a clear role of a demonstration activity in an R&D project
can be identified.
R&D in the social sciences, humanities and the arts

2.102 In the definition of R&D in this manual, the phrase “knowledge
of humankind, culture and society” includes the social sciences, humanities
and the arts. Also for the social sciences, humanities and arts, the use of clear
criteria to identify R&D, such as having an appreciable element of novelty
and dealing with uncertainty, is extremely helpful for defining the boundary
between R&D and related (routine) scientific activities as well as non-scientific
investigations. The conceptual, methodological and empirical components of
the project concerned have to be taken into consideration to identify an R&D
activity.
2.103 In the social sciences – e.g. sociology, economics or political science –
data collection activities, e.g. statistical surveys on specific populations, can
only be included in R&D if they are undertaken as an integral part of a specific
research project or for the benefit of a specific research project. Therefore,
projects of a routine nature, in which social scientists bring established
social science methodologies, principles and models to bear on a particular
problem, cannot be classified as research. For example, a project using labour
force survey data to identify long-term unemployment trends should exclude
the data collection activity as an R&D component (as those data are regularly
collected by using an existing methodology). On the other hand, a case-
study on unemployment in a specific region, if applying original techniques
in interviewing survey respondents could include such data collection in its
R&D effort. From a broader perspective, to the extent that the social sciences
are using empirical data, the same guidelines have to be applied as for the
natural sciences (although excluding the testing of their results on an
experimental basis).
2.104 For the humanities, the same approach could be used as for the arts
(studies on literature, music, visual arts, theatre, dance and other performing
arts). Their historical or comparative nature can be pointed out as well as the
relevant role played by universities or other specialised institutions in developing
scientific guidelines to be followed by the scholars in the field.

2.105 The broad range of sources used in history, archaeology, languages

and legal studies and the different methods used by researchers are possible
areas of R&D. The adoption of the five core criteria for R&D should be
recommended, mainly with reference to novelty, creativity and transferability
and/or reproducibility.
2.106 In the fields of philosophy and religious studies, for instance,
historical and comparative studies undertaken in line with current academic
standards are included in R&D. As a general rule cannot be given, beyond that
of strictly applying the R&D criteria, the use of the institutional approach is
also recommended (i.e. potentially excluding research-related activities on
philosophy and religion conducted outside recognised research institutions).
2.107 In conclusion, research in the humanities and the arts can be
included in R&D in so far as their own internal requirements for identifying the
“scientific” nature of such research are met. Additional practical guidelines follow.
● Context of performance (institutional criterion). Research carried out within
the framework of a university or an officially recognised research institution
(including museums, libraries, etc.) can be included in R&D.
● Adoption of recognised procedures. Research requires formalisation, and this
applies to the humanities. Research activities could be identified and their
results made available to the scientific community through their publication in
scientific journals. In so far as these features can be identified and a scientific
community is actively developing some rules to identify its own members, the
same rules can be applied for identifying R&D performance.
● Research in the humanities may deal with the systematic development of
theories or interpretations of texts, events, material remains or any other
available evidence. By convention, research activities carried out outside the
fields of R&D listed in Chapter 3 have to be excluded from R&D.
R&D and traditional knowledge

2.108 A cross-cutting research domain with some overlap with the
humanities and medicine is that dealing with “traditional knowledge”. Traditional
knowledge has been defined to be a largely tacit “cumulative body of knowledge,
know-how, practices and representations maintained and developed by peoples
with extended histories of interaction with the natural environment […]
a cultural complex that encompasses language, naming and classification
systems, resource use practices, ritual, spirituality and worldview” (ICSU and
UNESCO, 2002: 9). The issue of the relationship between traditional knowledge
and R&D is particularly relevant in developing countries where the existence of a
valuable stock of traditional knowledge can be a powerful incentive for domestic
and foreign organisations to set up R&D activities.
2.109 As a general rule, where activities associated with traditional
knowledge form part of an R&D project, the effort (financial and in terms of
human resources) should be counted as R&D. Otherwise they should be excluded.

75
Examples of different types of activities involving traditional knowledge that

should be counted as contributing to R&D are as follows:
● An R&D project may entail a scientific-based approach to establishing the
content of traditional knowledge, in disciplines such as ethno-science (ethno-
botany, ethno-pedology, ethno-forestry, ethno-veterinary medicine, and
ethno-ecology) or cognitive anthropology. In this case, R&D methods within
established disciplines are used to study traditional knowledge.
● The application of scientific methods to identify the active ingredient of local
health remedies and/or their effectiveness for certain medical conditions. In
this case, R&D methods are applied directly to traditional knowledge products
with the purpose of expanding the stock of scientific knowledge.
● Activities undertaken by traditional knowledge practitioners to expand the
stock of traditional knowledge, through the combined use of traditional and
other, scientific methods. These activities must meet the standard criteria for
being countable as R&D or must be carried out in universities.
2.110 Examples of traditional knowledge activities that would be excluded
from R&D include the following:
● the regular/continued use of traditional knowledge by practitioners, for
example, in treating ailments or managing crops
● the routine development of products based on traditional knowledge
● the storage and communication of traditional knowledge in traditional ways

(by the test of novelty)
● the traditional handing down of religious or cultural beliefs and practices.
2.8. Activities to be excluded from R&D

2.111 For survey purposes, R&D must be distinguished from a wide range of
related activities with a scientific and technological basis. These other activities
are very closely linked to R&D both through flows of information and in terms
of operations, institutions and personnel, but as far as possible they should be
excluded when measuring R&D.
Scientific and technical information services

2.112 The specialised activities of:
● collecting
● coding
● recording
● classifying
● disseminating
● translating

● analysing
● evaluating
by:
● scientific and technical personnel
● bibliographic services
● patent services
● scientific and technical information, extension and advisory services
● scientific conferences
are to be excluded, except when conducted solely or primarily for the

purpose of R&D support (e.g. the preparation of the original report of R&D
findings should be included in R&D) or in the context of R&D projects, as defined
earlier in this section.
Testing and standardisation

2.113 This concerns the maintenance of national standards, the calibration
of secondary standards and the routine testing and analysis of materials,
components, products, processes, soils, atmosphere, etc. These activities are not
R&D.
Feasibility studies
2.114 The investigation of proposed engineering projects, using existing
techniques to provide additional information before deciding on implementation,
is not R&D. In the social sciences, feasibility studies are investigations of the
socio-economic characteristics and implications of specific situations (e.g. a
study of the viability of a petrochemical complex in a certain region). However,
feasibility studies on research projects are part of R&D.
Specialised health care

2.115 This concerns the routine investigation and normal application of
specialised medical knowledge. Usually this is not R&D; there may, however, be
an element of R&D in what is usually called “specialised health care”, when it is
carried out, for example, in university hospitals.
Policy-related studies
2.116 In this context, “policy” refers not only to national policy but also to
policy at regional and local levels, as well as the policy of business enterprises
in the pursuit of their economic activity. Policy-related studies cover a range
of activities, such as the analysis and assessment of the existing programmes,
policies and operations of government departments and other institutions; the
work of units concerned with the continuing analysis and monitoring of external

77
Chapter Three
Types and Patterns

of Innovation
Tesla Motors
In 2015, Tesla Motors was a $3.2 billion company on track to set history. It had
created two cars that most people agreed were remarkable. Consumer reports
had rated Tesla’s Model S the best car it had ever reviewed. Though it was not
yet posting profits, sales were growing rapidly and analysts were hopeful that
profits would soon follow. It had repaid its government loans ahead of the major
auto conglomerates. Most importantly, it looked like it might survive. Perhaps
even thrive. This was astonishing as there had been no other successful auto
manufacturing start-up in the United States since the 1920s.
The road leading up to Tesla’s position in 2015 had not always been smooth,
and there were many doubts that still lingered. Tesla had benefited from the
enthusiasm of the “eco-wealthy”—a rather narrow portion of the market. How
would Tesla fare when it was in direct competition with General Motors, Ford,
and Nissan for the mass market? Would it be able to turn a sustainable profit
on its auto-making operations? Furthermore, some questioned whether Tesla’s
goals to sell to the mass market even made sense. In the niche market, it had a
privileged position with customers that were relatively price-insensitive and were
seeking a stylish, high-performance car that made an environmental statement.
To compete for the mass market, the car would have to provide good value for
the money (involving trade-offs that might conflict with Chairman Elon Musk’s
ideals), and the obstacles to charging would have to be overcome.
History of Tesla
In the year 2003, an engineer named Martin Eberhard was looking for his next
big project. A tall, slim man with a mop of gray hair, Eberhard was a serial entre-
preneur who had launched a number of start-ups, including a company called
NuvoMedia, which he sold to Gemstar in a $187 million deal. Eberhard was also
looking for a sports car that would be environmentally friendly—he had concerns
about global warming and U.S. dependence on the Middle East for oil. When he
didn’t find the car of his dreams on the market he began contemplating building
43
44 Part One Industry Dynamics of Technological Innovation
one himself, even though he had zero experience in the auto industry. Eberhard
noticed that many of the driveways that had a Toyota Prius hybrid electric
vehicle (or “dork mobile” as he called it) also had expensive sports cars in them—
making Eberhard speculate that there could be a market for a high-performance
environmentally friendly car. As explained by Eberhard, “It was clear that people
weren’t buying a Prius to save money on gas—gas was selling close to inflation–
adjusted all-time lows. They were buying them to make a statement about the
environment.”a
Eberhard began to consider a range of alternative fuel options for his car:
hydrogen fuel cells, natural gas, and diesel. However, he soon concluded that
the highest efficiency and performance would come from a pure electric vehicle.
Luckily for Eberhard, Al Cocconi (founder of AC Propulsion and one of the original
engineers for GM’s ill-fated EV-1) had concluded the same thing and had pro-
duced a car called the tzero. The tzero could go from zero to 60 miles per hour in
4.1 seconds, but it was powered with extremely heavy lead-acid batteries, limit-
ing its range to about 60 miles between charges. Eberhard approached Cocconi
with the idea of using the lighter lithium ion batteries, which offered six times
more energy per pound. Cocconi was eager to try out the idea (he had, in fact,
been experimenting with lithium ion batteries himself), and the resulting lithium
ion-based tzero accelerated to 60 miles per hour in 3.6 seconds, and could travel
more than 300 miles. Eberhard licensed the electric-drive-train technology from
AC Propulsion, and founded his company, Tesla Motors (named after Nikola Tesla,
a late nineteenth-century and early twentieth-century inventor who developed,
among other things, the AC electrical systems used in the United States today).b
Meanwhile, there was another entrepreneur—one with much deeper pockets—
also interested in developing electric vehicles based on the tzero: Elon Musk. In
2002, Elon Musk was a 31-year-old South African living in California, who had
founded a company that ultimately became PayPal. After selling PayPal to eBay
in 2002 for $1.5 billion, he started a company called SpaceX with the ambitious
goal of developing cheap, consumer space travel. (SpaceX’s Dragon spacecraft
ultimately made history in May of 2012 by becoming the first commercial vehicle
to launch and dock at the International Space Station.c) Musk was also the chair-
man of a high profile clean tech venture in Northern California called Solar City.
Musk’s assertive style, and his astonishing record of high-tech entrepreneurship,
made him one of the inspirations for the Tony Stark character in Jon Favreau’s
Iron Man movies.
Like Eberhard, Musk thought electric cars were the key to the United States
achieving energy independence, and he approached Cocconi about buying the
tzero. Tom Gage, who was then AC Propulsion’s CEO, suggested that Musk
collaborate with Eberhard. After a two hour meeting in February of 2004, Musk
agreed to fund Eberhard’s plan with $6.3 million. He would be the company’s
chairman and Eberhard would serve as CEO.
The Roadster
The first Tesla prototype, named the Roadster, was based on the $45,000 Lotus
Elise, a fast and light sports car that seemed perfect for the creation of Eberhard
Chapter 3 Types and Patterns of Innovation 45
and Musk’s grand idea. The car would have 400 volts of electric potential, liquid-
cooled lithium ion batteries, and a series of silicon transistors that would give the
car acceleration so powerful the driver would be pressed back against their seat.d
It would be about as fast as a Porsche 911 Turbo, would not create a single emis-
sion, and would get about 220 miles on a single charge from the kind of outlet
you would use to power a washing machine.e
After a series of clashes between Musk and Eberhard that led to delays in
launching the Roadster, Eberhard was pushed out of the company. The Roadster
missed its deadline for beginning production at the Lotus facility, triggering a
penalty built into the manufacturing contract Eberhard had signed with Lotus: a
$4 million fee. However, when the car finally launched in 2008, the enthusias-
tic response it received was astonishing—it boasted an all-star list of celebrities
with reservations to buy, and everywhere the Roadster drove, people stopped
to stare.f
The Model S
Musk’s ambitions did not stop at a niche high-end car. He wanted to build a major
U.S. auto company—a feat that had not been successfully accomplished since
the 1920s. To do so, he knew he needed to introduce a less-expensive car that
could attract a higher volume of sales, if not quite the mass market. In June of
2008, Tesla announced the Model S—a high-performance all-electric sedan that
would sell for a price ranging from $57,400 to $77,400 and compete against
cars like the BMW 5-series. The car would have an all-aluminum body, and a
range of up to 300 miles per charge.g The Model S cost $500 million to develop,h
however offsetting that cost was a $465 million loan Tesla received from the
U.S. government to build the car, as part of the U.S. government’s initiative to
promote the development of technologies that would help the United States to
achieve energy independence.
By May of 2012, Tesla reported that it already had 10,000 reservations for
customers hoping to buy the Model S, and Musk confidently claimed that the
company would soon be producing—and selling—20,000 Model S cars a year.
Musk also noted that after ramping up production, he expected to see “at least
10,000 units a year from demand in Europe and at least 5,000 in Asia.”i The
production of the Model S went more smoothly than that of the Roadster, and
by June of 2012, the first Model S cars were rolling off the factory floor. The very
first went to Jeff Skoll, eBay’s first president, and a major investor in Tesla. On the
day of the launch, Skoll talked with Musk about whether it was harder to build a
rocket or a car (referring to Musk’s SpaceXcompany): “We decided it was a car.
There isn’t a lot of competition in space.”j
To build the car, Tesla bought a recently closed automobile factory in Fremont,
California, that had been used for the New United Motor Manufacturing Inc.
(NUMMI) venture between Toyota and General Motors. The factory, which was
capable of producing 1,000 cars a week, was far bigger than Tesla’s immedi-
ate needs and would give the company room to grow. Furthermore, though
the plant and the land it was on had been appraised at around $1 billion
before NUMMI was shut down, Tesla was able to snap up the idled factory for
$42 million.k Tesla also used the factory to produce battery packs for Toyota’s
RAV4, and a charger for a subcompact Daimler AG electric vehicle. These proj-
ects would supplement Tesla’s income while also helping it to build scale and
learning curve efficiencies in its technologies.
In the first quarter of 2013, Tesla announced its first quarterly profit. The
company had taken in $562 million in revenues and reported an $11.2 million
profit. Then more good news came: The Model S had earned Consumer Reports’
highest rating and had outsold similarly priced BMW and Mercedes models in
the first quarter.l In May of 2013, the company raised $1 billion by issuing new
shares and then surprised investors by announcing that it had paid back its gov-
ernment loan. After repaying the loan, Tesla had about $679 million in cash.
Musk had announced confidently that he felt it was his obligation to pay back
taxpayer money as soon as possible and that the company had sufficient funds
now to develop its next generation of automobiles without the loan and without
issuing further shares.m
The Future of Tesla

By 2015, Tesla Motors was also in the process of developing a sport utility vehicle
that seats seven, the Model X, which cost $250 million to develop and would
be available in 2016.n This car was part of Musk’s longer-term ambition to tap a
more mainstream market for the cars.
Though Tesla’s moves had been bold and risky, its success thus far was inspir-
ing. The company had survived its infancy, appeared to be solvent, and was
meeting its sales objectives even though serious obstacles remained for electric
vehicles. It was also competing against companies with far greater scale. As
noted by O’Dell, a senior editor at auto information site Edmunds.com, on Tesla’s
success, “A lot of people have been very, very skeptical . . . when you want to
be an automaker, you are competing with multibillion-dollar conglomerates . . .
It’s entrepreneurism on steroids . . . They had a huge learning curve but they’ve
powered through it.” Theo O’Neill, an analyst at Wunderlich Securities adds that
“It’s going to prove everybody in Detroit wrong . . . They all say what Tesla is
doing isn’t possible.”o
Discussion Questions
1. Is the Tesla Model S a radical innovation or an incremental innovation?
Competence enhancing or destroying, and from whose perspective? Is it a
component or an architectural innovation?
2. What factors do you think influence the rate at which consumers have
adopted (or will adopt) the Tesla Model S?
3. Where do you think electric vehicle battery technology is on the technology
s curve?
4. Do you think Tesla Motors will be profitable? Why or why not?
a
Copeland, M.V. 2008. Tesla’s wild ride. Fortune, Vol. 158, issue 2, pg. 82–94.
b
c
Boudreau. J. 2012. In a Silicon Valley milestone, Tesla Motors begins delivering Model S electric cars.
June 24: Breaking News Section.
d
e
Williams, A. 2009. Taking a Tesla for a status check in New York. New York Times, July 19th, ST.7.
f
Williams, A. 2009. Taking a Tesla for a status check in New York. New York Times, July 19th, ST.7.
g
Ramsey, M. 2011. Tesla sets 300-mile range for second electric car. Wall Street Journal (Online),
March 7th: n/a
h
Vance, A. 2015. Elon Musk: Tesla, SpaceX, and the Quest for a Fantastic Future. New York:
Harper Collins
i
Sweet, C. 2013. Tesla posts its first quarterly profit. Wall Street Journal (Online), May 9th: n/a.
j
Oakland Tribune, June 24: Breaking News Section.
k
Anonymous. 2010. Idle Fremont plant gears up for Tesla. Wall Street Journal (Online),
October 20th: n/a
l
Levi, M. 2013. How Tesla pulled ahead of the electric-car pack. Wall Street Journal, June 21:A.11.
m
White, J.B. 2013. Corporate News: Electric car startup Tesla repays U.S. loan. Wall Street Journal, May
23rd:B.3.
n
Caranddriver.com, accessed May 11, 2015.
o
Oakland Tribune, June 24: Breaking News Section.
OVERVIEW
The previous chapters pointed out that technological innovation can come from many
sources and take many forms. Different types of technological innovations offer dif-
ferent opportunities for organizations and society, and they pose different demands
upon producers, users, and regulators. While there is no single agreed-upon taxonomy
to describe different kinds of technological innovations, in this chapter we will review
several dimensions that are often used to categorize technologies. These dimensions are
useful for understanding some key ways that one innovation may differ from another.
technology The path a technology follows through time is termed its technology trajectory.
trajectory Technology trajectories are most often used to represent the technology’s rate of per-
The path a tech- formance improvement or its rate of adoption in the marketplace. Though many fac-
nology takes
through its life-
tors can influence these technology trajectories (as discussed in both this chapter and
time. This path the following chapters), some patterns have been consistently identified in technology
may refer to its trajectories across many industry contexts and over many periods. Understanding
rate of perfor- these patterns of technological innovation provides a useful foundation that we will
mance improve- build upon in the later chapters on formulating technology strategy.
ment, its rate
of diffusion, or
The chapter begins by reviewing the dimensions used to distinguish types of
other change of innovations. It then describes the s-curve patterns so often observed in both the rate
interest. of technology improvement and the rate of technology diffusion to the market. In the
last section, the chapter describes research suggesting that technological innovation
follows a cyclical pattern composed of distinct and reliably occurring phases.
TYPES OF INNOVATION
Technological innovations are often described using dimensions such as “radical” ver-
sus “incremental.” Different types of innovation require different kinds of underlying
knowledge and have different impacts on the industry’s competitors and customers.
Four of the dimensions most commonly used to categorize innovations are described
here: product versus process innovation, radical versus incremental, competence enhanc-
ing versus competence destroying, and architectural versus component.
Product Innovation versus Process Innovation

Product innovations are embodied in the outputs of an organization—its goods or ser-
vices. For example, Honda’s development of a new hybrid electric vehicle is a product
innovation. Process innovations are innovations in the way an organization conducts its
business, such as in the techniques of producing or marketing goods or services. Process
innovations are often oriented toward improving the effectiveness or efficiency of pro-
duction by, for example, reducing defect rates or increasing the quantity that may be pro-
duced in a given time. For example, a process innovation at a biotechnology firm might
entail developing a genetic algorithm that can quickly search a set of disease-related
genes to identify a target for therapeutic intervention. In this instance, the process
innovation (the genetic algorithm) can speed up the firm’s ability to develop a product
innovation (a new therapeutic drug).
New product innovations and process innovations often occur in tandem. First, new
processes may enable the production of new products. For example, as discussed later
in the chapter, the development of new metallurgical techniques enabled the development
of the bicycle chain, which in turn enabled the development of multiple-gear bicycles.
Second, new products may enable the development of new processes. For example, the
development of advanced workstations has enabled firms to implement computer-aided
manufacturing processes that increase the speed and efficiency of production. Finally,
a product innovation for one firm may simultaneously be a process innovation for
another. For example, when United Parcel Service (UPS) helps a customer develop
a more efficient distribution system, the new distribution system is simultaneously a
product innovation for UPS and a process innovation for its customer.
Though product innovations are often more visible than process innovations, both
are extremely important to an organization’s ability to compete. Throughout the
radical remainder of the book, the term innovation will be used to refer to both product and
innovation process innovations.
An innovation
that is very new Radical Innovation versus Incremental Innovation
and different from
prior solutions.
One of the primary dimensions used to distinguish types of innovation is the continuum
between radical versus incremental innovation. A number of definitions have been posed
incremental for radical innovation and incremental innovation, but most hinge on the degree
innovation to which an innovation represents a departure from existing practices.1 Thus radicalness
An innovation might be conceived as the combination of newness and the degree of differentness. A
that makes a technology could be new to the world, new to an industry, new to a firm, or new merely
relatively minor
change from (or to an adopting business unit. A technology could be significantly different from existing
adjustment to) products and processes or only marginally different. The most radical innovations would
existing practices. be new to the world and exceptionally different from existing products and processes.
The introduction of wireless telecommunication products aptly illustrates this—it

embodied significantly new technologies that required new manufacturing and service
processes. Incremental innovation is at the other end of the spectrum. An incremental
innovation might not be particularly new or exceptional; it might have been previously
known to the firm or industry, and involve only a minor change from (or adjustment to)
existing practices. For example, changing the configuration of a cell phone from one that
has an exposed keyboard to one that has a flip cover or offering a new service plan that
enables more free weekend minutes would represent incremental innovation.
The radicalness of innovation is also sometimes defined in terms of risk. Since radi-
cal innovations often embody new knowledge, producers and customers will vary in
their experience and familiarity with the innovation, and in their judgment of its use-
fulness or reliability.2 The development of third generation (3G) telephony is illustrative.
3G wireless communication technology utilizes broadband channels. This increased
bandwidth gives mobile phones far greater data transmission capabilities that enable
activities such as videoconferencing and accessing the most advanced Internet sites.
For companies to develop and offer 3G wireless telecommunications service required
a significant investment in new networking equipment and an infrastructure capable
of carrying a much larger bandwidth of signals. It also required developing phones
with greater display and memory capabilities, and either increasing the phone’s bat-
tery power or increasing the efficiency of the phone’s power utilization. Any of these
technologies could potentially pose serious obstacles. It was also unknown to what
degree customers would ultimately value broadband capability in a wireless device.
Thus, the move to 3G required managers to assess several different risks simultane-
ously, including technical feasibility, reliability, costs, and demand.
Finally, the radicalness of an innovation is relative, and may change over time or
with respect to different observers. An innovation that was once considered radical
may eventually be considered incremental as the knowledge base underlying the inno-
competence- vation becomes more common. For example, while the first steam engine was a monu-
enhancing mental innovation, today its construction seems relatively simple. Furthermore, an
(-destroying) innovation that is radical to one firm may seem incremental to another. Although both
innovation Kodak and Sony introduced digital cameras for the consumer market within a year of
An innovation
each other (Kodak’s DC40 was introduced in 1995, and Sony’s Cyber-Shot Digital
that builds on
(renders obsolete) Still Camera was introduced in 1996), the two companies’ paths to the introduction
existing knowl- were quite different. Kodak’s historical competencies and reputation were based on its
edge and skills. expertise in chemical photography, and thus the transition to digital photography and
Whether an inno- video required a significant redirection for the firm. Sony, on the other hand, had been
vation is compe-
an electronics company since its inception, and had a substantial level of expertise in
tence enhancing
or competence digital recording and graphics before producing a digital camera. Thus, for Sony, a
destroying digital camera was a straightforward extension of its existing competencies.
depends on
whose perspec-
tive is being
Competence-Enhancing Innovation versus
taken. An Competence-Destroying Innovation
innovation can Innovations can also be classified as competence enhancing versus competence
be competence
enhancing to one
destroying. An innovation is considered to be competence enhancing from the
firm, while com- perspective of a particular firm if it builds on the firm’s existing knowledge base. For
petence destroy- example, each generation of Intel’s microprocessors (e.g., 286, 386, 486, Pentium,
ing for another. Pentium II, Pentium III, Pentium 4) builds on the technology underlying the previous
generation. Thus, while each generation embodies innovation, these innovations lever-
age Intel’s existing competencies, making them more valuable.
An innovation is considered to be competence destroying from the perspective of a
particular firm if the technology does not build on the firm’s existing competencies or
renders them obsolete. For example, from the 1600s to the early 1970s, no self-respecting
mathematician or engineer would have been caught without a slide rule. Slide rules are
lightweight devices, often constructed of wood, that use logarithm scales to solve com-
plex mathematical functions. They were used to calculate everything from the structural
properties of a bridge to the range and fuel use of an aircraft. Specially designed slide
rules for businesses had, for example, scales for doing loan calculations or determin-
ing optimal purchase quantities. During the 1950s and 1960s, Keuffel & Esser was the
preeminent slide-rule maker in the United States, producing 5,000 slide rules a month.
However, in the early 1970s, a new innovation relegated the slide rule to collectors and
museum displays within just a few years: the inexpensive handheld calculator. Keuffel
& Esser had no background in the electronic components that made electronic calcula-
tors possible and was unable to transition to the new technology. By 1976, Keuffel &
Esser withdrew from the market.3 Whereas the inexpensive handheld calculator built on
the existing competencies of companies such as Hewlett-Packard and Texas Instruments
(and thus for them would be competence enhancing), for Keuffel & Esser, the calculator
was a competence-destroying innovation.
Architectural Innovation versus Component Innovation

Most products and processes are hierarchically nested systems, meaning that at any
unit of analysis, the entity is a system of components, and each of those components is,
in turn, a system of finer components, until we reach some point at which the com-
ponents are elementary particles.4 For example, a bicycle is a system of components
such as a frame, wheels, tires, seat, brakes, and so on. Each of those components is also
component
a system of components: The seat might be a system of components that includes a
(or modular)
innovation
metal and plastic frame, padding, a nylon cover, and so on.
An innovation to An innovation may entail a change to individual components, to the overall archi-
one or more com- tecture within which those components operate, or both. An innovation is considered
ponents that does a component innovation (or modular innovation) if it entails changes to one
not significantly or more components, but does not significantly affect the overall configuration of the
affect the overall
configuration of
system.5 In the example above, an innovation in bicycle seat technology (such as the
the system. incorporation of gel-filled material for additional cushioning) does not require any
changes in the rest of the bicycle architecture.
architectural In contrast, an architectural innovation entails changing the overall design of
innovation the system or the way that components interact with each other. An innovation that
An innovation is strictly architectural may reconfigure the way that components link together in the
that changes the
overall design
system, without changing the components themselves.6 Most architectural innova-
of a system or tions, however, create changes in the system that reverberate throughout its design,
the way its com- requiring changes in the underlying components in addition to changes in the ways
ponents interact those components interact. Architectural innovations often have far-reaching and
with each other. complex influences on industry competitors and technology users.
For example, the transition from the high-wheel bicycle to the safety bicycle was
an architectural innovation that required (and enabled) the change of many compo-
nents of the bicycle and the way in which riders propelled themselves. In the 1800s,
bicycles had extremely large front wheels. Because there were no gears, the size of
the front wheel directly determined the speed of the bicycle since the circumference
of the wheel was the distance that could be traveled in a single rotation of the ped-
als. However, by the start of the twentieth century, improvements in metallurgy had
enabled the production of a fine chain and a sprocket that was small enough and light
enough for a human to power. This enabled bicycles to be built with two equally sized
wheels, while using gears to accomplish the speeds that the large front wheel had
enabled. Because smaller wheels meant shorter shock-absorbing spokes, the move to
smaller wheels also prompted the development of suspension systems and pneumatic
(air-filled) tires. The new bicycles were lighter, cheaper, and more flexible. This
architectural innovation led to the rise of companies such as Dunlop (which invented
the pneumatic tire) and Raleigh (which pioneered the three-speed, all-steel bicycle),
and transformed the bicycle from a curiosity into a practical transportation device.
For a firm to initiate or adopt a component innovation may require that the firm
have knowledge only about that component. However, for a firm to initiate or adopt an
architectural innovation typically requires that the firm have architectural knowledge
about the way components link and integrate to form the whole system. Firms must
be able to understand how the attributes of components interact, and how changes in
some system features might trigger the need for changes in many other design features
of the overall system or the individual components.
Using the Dimensions

Though the dimensions described above are useful for exploring key ways that one
innovation may differ from another, these dimensions are not independent, nor do they
offer a straightforward system for categorizing innovations in a precise and consistent
manner. Each of the above dimensions shares relationships with others—for example,
architectural innovations are often considered more radical and more competence
destroying than component innovations. Furthermore, how an innovation is described
on a dimension often depends on who is doing the describing and with what it is being
compared. An all-electric vehicle, for example, might seem like a radical and compe-
tence destroying innovation to a manufacturer of internal combustion engines, but to a
customer who only has to change how they fuel/charge the vehicle, it might seem like
an incremental and competence-enhancing innovation. Thus, while the dimensions
above are valuable for understanding innovation, they should be considered relative
dimensions whose meaning is dependent on the context in which they are used.
We now will turn to exploring patterns in technological innovation. Numerous
studies of innovation have revealed recurring patterns in how new technologies
emerge, evolve, are adopted, and are displaced by other technologies. We begin by
examining technology s-curves.
TECHNOLOGY S-CURVES
Both the rate of a technology’s performance improvement and the rate at which the
technology is adopted in the marketplace repeatedly have been shown to conform to an
s-shape curve. Though s-curves in technology performance and s-curves in technology
diffusion are related (improvements in performance may foster faster adoption, and
greater adoption may motivate further investment in improving performance), they are
fundamentally different processes. S-curves in technology improvement are described

first, followed by s-curves in technology diffusion. This section also explains that
despite the allure of using s-curves to predict when new phases of a technology’s life
cycle will begin, doing so can be misleading.
S-Curves in Technological Improvement

Many technologies exhibit an s-curve in their performance improvement over their
lifetimes.7 When a technology’s performance is plotted against the amount of effort
and money invested in the technology, it typically shows slow initial improvement,
then accelerated improvement, then diminishing improvement (see Figure 3.1).
Performance improvement in the early stages of a technology is slow because the
fundamentals of the technology are poorly understood. Great effort may be spent
exploring different paths of improvement or different drivers of the technology’s
improvement. If the technology is very different from previous technologies, there
may be no evaluation routines that enable researchers to assess its progress or its poten-
tial. Furthermore, until the technology has established a degree of legitimacy, it may
be difficult to attract other researchers to participate in its development.8 However, as
scientists or firms gain a deeper understanding of the technology, improvement begins
to accelerate. The technology begins to gain legitimacy as a worthwhile endeavor,
attracting other developers. Furthermore, measures for assessing the technology are
developed, permitting researchers to target their attention toward those activities that
reap the greatest improvement per unit of effort, enabling performance to increase
rapidly. However, at some point, diminishing returns to effort begin to set in. As the
technology begins to reach its inherent limits, the cost of each marginal improvement
increases, and the s-curve flattens.
Often a technology’s s-curve is plotted with performance (e.g., speed, capacity, or
power) against time, but this must be approached with care. If the effort invested is not
constant over time, the resulting s-curve can obscure the true relationship. If effort is
FIGURE 3.1
S-Curve of Limit of Technology
Technology
Performance
Performance
Effort
relatively constant over time, plotting performance against time will result in the same
characteristic curve as plotting performance against effort. However, if the amount of
effort invested in a technology decreases or increases over time, the resulting curve could
appear to flatten much more quickly, or not flatten at all. For instance, one of the more
well-known technology trajectories is described by an axiom that became known as
Moore’s law. In 1965, Gordon Moore, cofounder of Intel, noted that the density of tran-
sistors on integrated circuits had doubled every year since the integrated circuit was
invented. That rate has since slowed to doubling every 18 months, but the rate of accel-
eration is still very steep. Figure 3.2 reveals a sharply increasing performance curve.
However, Intel’s rate of investment (research and development dollars per year)
has also been increasing rapidly, as shown in Figure 3.3. Not all of Intel’s R&D
expense goes directly to improving microprocessor power, but it is reasonable to
assume that Intel’s investment specifically in microprocessors would exhibit a simi-
lar pattern of increase. Figure 3.3 shows that the big gains in transistor density have
come at a big cost in terms of effort invested. Though the curve does not yet resemble
the traditional s-curve, its rate of increase is not as sharp as when the curve is plotted
against years. Gordon Moore predicted that transistor miniaturization will reach its
discontinuous physical limits about 2017.
technology Technologies do not always get the opportunity to reach their limits; they may
A technology be rendered obsolete by new, discontinuous technologies. A new innovation is
that fulfills a discontinuous when it fulfills a similar market need, but does so by building on an
similar market entirely new knowledge base.9 For example, the switches from propeller-based planes
need by building
on an entirely to jets, from silver halide (chemical) photography to digital photography, from carbon
new knowledge copying to photocopying, and from vinyl records (or analog cassettes) to compact
base. discs were all technological discontinuities.
FIGURE 3.2
Improvements in Intel’s Microprocessor Transistor Density over Time
Year Transistors Intel CPU 800,000,000

1971 2,250 4004 700,000,000
1972 2,500 8008
1974 5,000 8080 600,000,000
Transistor Density
1978 29,000 8086

500,000,000
1982 120,000 286
1985 275,000 386™ 400,000,000
1989 1,180,000 486™ DX
1993 3,100,000 Pentium® 300,000,000
1997 7,500,000 Pentium II 200,000,000
1999 24,000,000 Pentium III
2000 42,000,000 Pentium 4 100,000,000
2002 55,000,000 Pentium M
-
2003 220,000,000 Itanium 2 1970 1975 1980 1985 1990 1995 2000 2005 2010
2005 291,000,000 Pentium D
Year
2006 582,000,000 Core 2 Quad
2007 731,000,000 Core i7 (Quad)
FIGURE 3.3
800,000,000
Graph of
Transistor
700,000,000
Density versus
Cumulative
600,000,000
R&D Expense,
1972–2007
Transistor Density 500,000,000
400,000,000
300,000,000
200,000,000
100,000,000
0
0 10,000 20,000 30,000 40,000 50,000 60,000
Cumulative R&D Expense ($millions)
Initially, the technological discontinuity may have lower performance than the
incumbent technology. For instance, one of the earliest automobiles, introduced in 1771
by Nicolas Joseph Cugnot, was never put into commercial production because it was
much slower and harder to operate than a horse-drawn carriage. It was three-wheeled,
steam-powered, and could travel at 2.3 miles per hour. A number of steam- and gas-
powered vehicles were introduced in the 1800s, but it was not until the early 1900s
that automobiles began to be produced in quantity.
In early stages, effort invested in a new technology may reap lower returns than
effort invested in the current technology, and firms are often reluctant to switch.
However, if the disruptive technology has a steeper s-curve (see Figure 3.4a) or an
s-curve that increases to a higher performance limit (see Figure 3.4b), there may come
a time when the returns to effort invested in the new technology are much higher
than effort invested in the incumbent technology. New firms entering the industry
are likely to choose the disruptive technology, and incumbent firms face the difficult
choice of trying to extend the life of their current technology or investing in switch-
ing to the new technology. If the disruptive technology has much greater performance
potential for a given amount of effort, in the long run it is likely to displace the incum-
bent technology, but the rate at which it does so can vary significantly.
S-Curves in Technology Diffusion

S-curves are also often used to describe the diffusion of a technology. Unlike s-curves
technology in technology performance, s-curves in technology diffusion are obtained by plot-
diffusion ting the cumulative number of adopters of the technology against time. This yields
The spread of an s-shape curve because adoption is initially slow when an unfamiliar technology is
a technology
through a
introduced to the market; it accelerates as the technology becomes better understood
population. and utilized by the mass market, and eventually the market is saturated so the rate of
new adoptions declines. For instance, when electronic calculators were introduced to
FIGURE 3.4 the market, they were first adopted by the

Technology relatively small pool of scientists and
S-Curves— Second engineers. This group had previously used
Introduction of technology
slide rules. Then the calculator began to
Discontinuous Performance penetrate the larger markets of accoun-
Technology First tants and commercial users, followed by
technology
the still larger market that included stu-
dents and the general public. After these
markets had become saturated, fewer
Effort opportunities remained for new adoptions.10
(a) One rather curious feature of technology
diffusion is that it typically takes far more
time than information diffusion.11 For
example, Mansfield found that it took
First 12 years for half the population of poten-
Performance technology tial users to adopt industrial robots, even
Second
technology
though these potential users were aware
of the significant efficiency advantages
the robots offered.12 If a new technology
is a significant improvement over existing
Effort solutions, why do some firms shift to it
(b) more slowly than others? The answer may
lie in the complexity of the knowledge
underlying new technologies, and in the
development of complementary resources that make those technologies useful. Although
some of the knowledge necessary to utilize a new technology might be transmitted
through manuals or other documentation, other aspects of knowledge necessary to fully
realize the potential of a technology might be built up only through experience. Some of
the knowledge about the technology might be tacit and require transmission from person
to person through extensive contact. Many potential adopters of a new technology will
not adopt it until such knowledge is available to them, despite their awareness of the
technology and its potential advantages.13
Furthermore, many technologies become valuable to a wide range of potential users
only after a set of complementary resources are developed for them. For example,
while the first electric light was invented in 1809 by Humphry Davy, an English
chemist, it did not become practical until the development of bulbs within which the
arc of light would be encased (first demonstrated by James Bowman Lindsay in 1835)
and vacuum pumps to create a vacuum inside the bulb (the mercury vacuum pump
was invented by Herman Sprengel in 1875). These early lightbulbs burned for only a
few hours. Thomas Alva Edison built on the work of these earlier inventors when, in
1880, he invented filaments that would enable the light to burn for 1,200 hours. The
role of complementary resources and other factors influencing the diffusion of tech-
nological innovations are discussed further in Chapters 4, 5, and 13.
Finally, it should be clear that the s-curves of diffusion are in part a function of the
s-curves in technology improvement: As technologies are better developed, they become
more certain and useful to users, facilitating their adoption. Furthermore, as learning-
curve and scale advantages accrue to the technology, the price of finished goods often
FIGURE 3.5
Average $1,000
Sales Prices
of Consumer $800
Electronics
Source: Consumer
$600
Electronics
Association.
$400
$200
$0
00
02
04
90
80
96
92
86
98
82
88
94
84
20
20
20
19
19
19
19
19
19
19
19
19
19
VCR CD Player Cell Phone
FIGURE 3.6
Penetration 100%
of Consumer 90%
Percent of U.S. Households
Electronics 80%
70%
Source: Consumer
Electronics 60%
Association. 50%
40%
30%
20%
10%
0%
00
02
04
90
80
96
92
86
98
82
88
94
84
20
20
20
19
19
19
19
19
19
19
19
19
19
VCR CD Player Cell Phone
drops, further accelerating adoption by users. For example, as shown in Figures 3.5
and 3.6, drops in average sales prices for video recorders, compact disc players, and
cell phones roughly correspond to their increases in household penetration.
S-Curves as a Prescriptive Tool

Several authors have argued that managers can use the s-curve model as a tool for pre-
dicting when a technology will reach its limits and as a prescriptive guide for whether
and when the firm should move to a new, more radical technology.14 Firms can use
data on the investment and performance of their own technologies, or data on the
overall industry investment in a technology and the average performance achieved by
multiple producers. Managers could then use these curves to assess whether a technol-
ogy appears to be approaching its limits or to identify new technologies that might be
emerging on s-curves that will intersect the firm’s technology s-curve. Managers could
then switch s-curves by acquiring or developing the new technology. However, as a
prescriptive tool, the s-curve model has several serious limitations.
Limitations of S-Curve Model as a Prescriptive Tool

First, it is rare that the true limits of a technology are known in advance, and there is
often considerable disagreement among firms about what a technology’s limits will be.
Second, the shape of a technology’s s-curve is not set in stone. Unexpected changes
in the market, component technologies, or complementary technologies can shorten or
extend the life cycle of a technology. Furthermore, firms can influence the shape of
the s-curve through their development activities. For example, firms can sometimes
stretch the s-curve through implementing new development approaches or revamping
the architecture design of the technology.15
Christensen provides an example of this from the disk-drive industry. A disk
drive’s capacity is determined by its size multiplied by its areal recording density;
thus, density has become the most pervasive measure of disk-drive performance.
In 1979, IBM had reached what it perceived as a density limit of ferrite-oxide–based
disk drives. It abandoned its ferrite-oxide–based disk drives and moved to developing
thin-film technology, which had greater potential for increasing density. Hitachi and
Fujitsu continued to ride the ferrite-oxide s-curve, ultimately achieving densities that
were eight times greater than the density that IBM had perceived to be a limit.
Finally, whether switching to a new technology will benefit a firm depends on a
number of factors, including (a) the advantages offered by the new technology, (b) the
new technology’s fit with the firm’s current abilities (and thus the amount of effort
that would be required to switch, and the time it would take to develop new competen-
cies), (c) the new technology’s fit with the firm’s position in complementary resources
(e.g., a firm may lack key complementary resources, or may earn a significant portion
of its revenues from selling products compatible with the incumbent technology), and
(d) the expected rate of diffusion of the new technology. Thus, a firm that follows an s-curve
model too closely could end up switching technologies earlier or later than it should.
TECHNOLOGY CYCLES
The s-curve model above suggests that technological change is cyclical: Each new
s-curve ushers in an initial period of turbulence, followed by rapid improvement, then
diminishing returns, and ultimately is displaced by a new technological discontinuity.16
The emergence of a new technological discontinuity can overturn the existing competi-
tive structure of an industry, creating new leaders and new losers. Schumpeter called
this process creative destruction, and argued that it was the key driver of progress in a
capitalist society.17
Several studies have tried to identify and characterize the stages of the technology
cycle in order to better understand why some technologies succeed and others fail,
and whether established firms or new firms are more likely to be successful in intro-
ducing or adopting a new technology.18 One technology evolution model that rose to
Research Brief
The Diffusion of Innovation and Adopter
Categories
S-curves in technology diffusion are often early adopters make excellent missionaries for new
explained as a process of different categories products or processes. Rogers estimated that the
of people adopting the technology at different next 13.5 percent of individuals to adopt an inno-
times. One typology of adopter categories that vation (after innovators) are in this category.
gained prominence was proposed by Everett
M. Rogers.a Figure 3.7 shows each of Rogers’s EARLY MAJORITY
adopter categories on a technology diffusion Rogers identifies the next 34 percent of individuals
s-curve. The figure also shows that if the non- in a social system to adopt a new innovation as the
cumulative share of each of these adopter groups early majority. The early majority adopts innova-
is plotted on the vertical axis with time on the tions slightly before the average member of a
horizontal axis, the resulting curve is typically social system. They are typically not opinion lead-
bell shaped (though in practice it may be skewed ers, but they interact frequently with their peers.
right or left).
LATE MAJORITY
INNOVATORS The next 34 percent of the individuals in a social
Innovators are the first individuals to adopt an system to adopt an innovation are the late major-
innovation. Extremely adventurous in their pur- ity, according to Rogers. Like the early majority, the
chasing behavior, they are comfortable with late majority constitutes one-third of the individ-
a high degree of complexity and uncertainty. uals in a social system. Those in the late majority
Innovators typically have access to substantial approach innovation with a skeptical air and may
financial resources (and thus can afford the losses not adopt the innovation until they feel pres-
incurred in unsuccessful adoption decisions). sure from their peers. The late majority may have
Though they are not always well integrated into scarce resources, thus making them reluctant to
a particular social system, innovators play an invest in adoption until most of the uncertainty
extremely important role in the diffusion of an about the innovation has been resolved.
innovation because they are the individuals who
bring new ideas into the social system. Rogers LAGGARDS
estimated that the first 2.5 percent of individuals The last 16 percent of the individuals in a social
to adopt a new technology are in this category. system to adopt an innovation are termed lag-
gards. They may base their decisions primarily
EARLY ADOPTERS
upon past experience rather than influence from
The second category of adopters is the early
the social network, and they possess almost no
adopters. Early adopters are well integrated into
opinion leadership. They are highly skeptical of
their social system and have the greatest poten-
innovations and innovators, and they must feel
tial for opinion leadership. Early adopters are
certain that a new innovation will not fail before
respected by their peers and know that to retain
adopting it.
that respect they must make sound innovation
adoption decisions. Other potential adopters look a
E. M. Rogers, Diffusion of Innovations, 3rd ed. (New
to early adopters for information and advice; thus York: Free Press, 1983).
continued
concluded
FIGURE 3.7
Technology Diffusion S-Curve with Adopter Categories
S-curve of Cumulative Adopters
100%
Cumulative
Adopters Laggards
84%
Late Majority
50%
Early Majority
16%
Early Adopters
2.5% Innovators
Time
(a)
Normal (Bell-Shaped) Curve of Market Share

Innovators Early Early Late Laggards
Adopters Majority Majority
34%
Share
13.5%
2.5%
Time
(b)
Theory in Action “Segment Zero”—A Serious Threat to Microsoft?
From 1980 to 2012, Microsoft was entrenched as the feature-rich products can command higher margins.
dominant personal computer operating system, giv- Though customers may also expect to have better-
ing it enormous influence over many aspects of the performing products over time, their ability to fully
computer hardware and software industries. Though utilize such performance improvements is slowed
competing operating systems had been introduced by the need to learn how to use new features and
during that time (e.g., Unix, Geoworks, NeXTSTEP, adapt their work and lifestyles. Thus, while both
Linux, and the Mac OS), Microsoft’s share of the per- the trajectory of technology improvement and the
sonal computer operating system market held sta- trajectory of customer demands are upward slop-
ble at roughly 85 percent throughout most of that ing, the trajectory for technology improvement is
period. In 2013, however, Microsoft’s dominance steeper (for simplicity, the technology trajectories
in computer operating systems was under greater are drawn in Figure 3.8 as straight lines and plotted
threat than it had ever been. A high-stakes race for against time in order to compare them against cus-
dominance over the next generation of computing tomer requirements).
was well underway, and Microsoft was not even in In Figure 3.8, the technology trajectory begins
the front pack. at a point where it provides performance close to
that demanded by the mass market, but over time
“SEGMENT ZERO” it increases faster than the expectations of the mass
As Andy Grove, former CEO of Intel, noted in market as the firm targets the high-end market. As
1998, in many industries—including microproces- the price of the technology rises, the mass market
sors, software, motorcycles, and electric vehicles— may feel it is overpaying for technological features it
technologies improve faster than customer demands does not value. In Figure 3.9, the low-end market is
of those technologies increase. Firms often add fea- not being served; it either pays far more for technol-
tures (speed, power, etc.) to products faster than cus- ogy that it does not need, or it goes without. It is this
tomers’ capacity to absorb them. Why would firms market that Andy Grove, former CEO of Intel, refers
provide higher performance than that required by to as segment zero.
the bulk of their customers? The answer appears to For Intel, segment zero was the market for low-
lie in the market segmentation and pricing objec- end personal computers (those less than $1,000).
tives of a technology’s providers. As competition in While segment zero may seem unattractive in terms
an industry drives prices and margins lower, firms of margins, if it is neglected, it can become the
often try to shift sales into progressively higher tiers breeding ground for companies that provide lower-
of the market. In these tiers, high performance and end versions of the technology. As Grove notes,
FIGURE 3.8 FIGURE 3.9

Trajectories of Technology Improvement and Low-End Technology’s Trajectory Intersects Mass
Customer Requirements Market Trajectory
Technology
High-end
trajectory
technology
High-end market High-end market
Performance Mass market Mass market

Performance
Low-end market Low-end market
Low-end
technology
Time Time
continued
60
concluded
“The overlooked, underserved, and seemingly business model used for the phones was also extremely
unprofitable end of the market can provide fertile attractive to both developers and customers, and
ground for massive competitive change.”a quickly resulted in enormous libraries of applications
As the firms serving low-end markets with sim- that ranged from the ridiculous to the indispensible.
pler technologies ride up their own trajectories From a traditional economics perspective, the
(which are also steeper than the slope of the trajec- phone operating system market should not be that
tories of customer expectations), they can eventually attractive to Microsoft—people do not spend as much
reach a performance level that meets the demands on the applications, and the carriers have too much
of the mass market, while offering a much lower bargaining power, among other reasons. However,
price than the premium technology (see Figure 3.9). those smartphone operating systems soon became
At this point, the firms offering the premium tech- tablet operating systems, and tablets were rapidly
nology may suddenly find they are losing the bulk becoming fully functional computers. Suddenly,
of their sales revenue to industry contenders that do all of that mindshare that Apple and Google had
not look so low end anymore. For example, by 1998, achieved in smartphone operating systems was
the combination of rising microprocessor power and transforming into mindshare in personal computer
decreasing prices enabled personal computers priced operating systems. Despite years of masterminding
under $1,000 to capture 20 percent of the market. the computing industry, Microsoft’s dominant posi-
tion was at risk of evaporating. The outcome was
THE THREAT TO MICROSOFT still uncertain–in 2015 Microsoft had an impressive
So where was the “segment zero” that could threaten arsenal of capital, talent, and relationships in its
Microsoft? Look in your pocket. In 2015, Apple’s armory—but for the first time, it was fighting the
iPhone operating system (iOS) and Google’s Android battle from a disadvantaged position.
collectively controlled over 90 percent of the world-
wide market for smartphone, followed by Research
in Motion’s Blackberry.b Gartner estimates put a
A. S. Grove, “Managing Segment Zero,” Leader to Leader,
Microsoft’s share at 3 percent. The iOS and Android 1999, p. 11.
interfaces offered a double whammy of beautiful aes- b
Dignan, L. 2013. Android, Apple iOS flip consumer, corpo-
thetics and remarkable ease of use. The applications rate market share. Between the Lines, February 13th.
prominence was proposed by Utterback and Abernathy. They observed that a tech-
nology passed through distinct phases. In the first phase (what they termed the fluid
phase), there was considerable uncertainty about both the technology and its market.
Products or services based on the technology might be crude, unreliable, or expensive,
but might suit the needs of some market niches. In this phase, firms experiment with
different form factors or product features to assess the market response. Eventually,
however, producers and customers begin to arrive at some consensus about the desired
dominant product attributes, and a dominant design emerges.19 The dominant design estab-
design lishes a stable architecture for the technology and enables firms to focus their efforts
A product design on process innovations that make production of the design more effective and effi-
that is adopted
by the majority
cient or on incremental innovations to improve components within the architecture.
of producers, Utterback and Abernathy termed this phase the specific phase because innovations
typically creating in products, materials, and manufacturing processes are all specific to the dominant
a stable archi- design. For example, in the United States the vast majority of energy production is
tecture on which based on the use of fossil fuels (e.g., oil, coal), and the methods of producing energy
the industry can
focus its efforts.
based on these fuels are well established. On the other hand, technologies that produce
energy based on renewable resources (e.g., solar, wind, hydrogen) are still in the fluid
61
FIGURE 3.10
The Era of Ferment
Technology Design Competition
Substitution
Cycle
Technological Dominant Design

Discontinuity Selected
Era of Incremental Change

Elaboration of Dominant Design
phase. Organizations such as Royal Dutch/Shell, General Electric, and Ballard Power
are experimenting with various forms of solar photocell technologies, wind-turbine
technologies, and hydrogen fuel cells in efforts to find methods of using renewable
resources that meet the capacity and cost requirements of serving large populations.
Building on the Utterback and Abernathy model, Anderson and Tushman studied the
history of the U.S. minicomputer, cement, and glass industries through several cycles
of technological change. Like Utterback and Abernathy, Anderson and Tushman found
that each technological discontinuity inaugurated a period of turbulence and uncertainty
(which they termed the era of ferment) (see Figure 3.10). The new technology might offer
breakthrough capabilities, but there is little agreement about what the major subsystems
of the technology should be or how they should be configured together. Furthermore, as
later researchers noted, during the era of ferment different stakeholders might have dif-
ferent concepts of what purpose the technology should serve, or how a business model
might be built around it.20 Thus, while the new technology displaces the old (Anderson
and Tushman refer to this as substitution), there is considerable design competition as
firms experiment with different forms of the technology. Just as in the Utterback and
Abernathy model, Anderson and Tushman found that a dominant design always arose
to command the majority of the market share unless the next discontinuity arrived too
soon and disrupted the cycle, or several producers patented their own proprietary tech-
nologies and refused to license to each other. Anderson and Tushman also found that the
dominant design was never in the same form as the original discontinuity, but it was also
never on the leading edge of the technology. Instead of maximizing performance on any
individual dimension of the technology, the dominant design tended to bundle together
a combination of features that best fulfilled the demands of the majority of the market.
In the words of Anderson and Tushman, the rise of a dominant design signals the
transition from the era of ferment to the era of incremental change.21 In this era, firms
focus on efficiency and market penetration. Firms may attempt to achieve greater market
segmentation by offering different models and price points. They may also attempt to
lower production costs by simplifying the design or improving the production process.
This period of accumulating small improvements may account for the bulk of the techno-
logical progress in an industry, and it continues until the next technological discontinuity.
Understanding the knowledge that firms develop during different eras lends insight
into why successful firms often resist the transition to a new technology, even if it
provides significant advantages. During the era of incremental change, many firms
cease to invest in learning about alternative design architectures and instead invest
in refining their competencies related to the dominant architecture. Most competition
revolves around improving components rather than altering the architecture; thus,
companies focus their efforts on developing component knowledge and knowledge
related to the dominant architecture. As firms’ routines and capabilities become more
and more wedded to the dominant architecture, the firms become less able to identify
and respond to a major architectural innovation. For example, the firm might estab-
lish divisions based on the primary components of the architecture and structure the
communication channels between divisions on the basis of how those components
interact. In the firm’s effort to absorb and process the vast amount of information avail-
able to it, it is likely to establish filters that enable it to identify the information most
crucial to its understanding of the existing technology design.22 As the firm’s exper-
tise, structure, communication channels, and filters all become oriented around maxi-
mizing its ability to compete in the existing dominant design, they become barriers to
the firm’s recognizing and reacting to a new technology architecture.
While many industries appear to conform to this model in which a dominant design
emerges, there are exceptions. In some industries, heterogeneity of products and
production processes are a primary determinant of value, and thus a dominant design
is undesirable.23 For example, art and cuisine may be examples of industries in which
there is more pressure to do things differently than to settle upon a standard.
Summary 1. Different dimensions have been used to distinguish types of innovation. Some of

of the most widely used dimensions include product versus process innovation, radical
versus incremental innovation, competence-enhancing versus competence-destroying
Chapter innovation, and architectural versus component innovation.
2. A graph of technology performance over cumulative effort invested often exhibits an
s-shape curve. This suggests that performance improvement in a new technology is
initially difficult and costly, but, as the fundamental principles of the technology are
worked out, it then begins to accelerate as the technology becomes better understood,
and finally diminishing returns set in as the technology approaches its inherent limits.
3. A graph of a technology’s market adoption over time also typically exhibits an
s-shape curve. Initially the technology may seem uncertain and there may be great
costs or risks for potential adopters. Gradually, the technology becomes more cer-
tain (and its costs may be driven down), enabling the technology to be adopted by
larger market segments. Eventually the technology’s diffusion slows as it reaches
market saturation or is displaced by a newer technology.
4. The rate at which a technology improves over time is often faster than the rate at
which customer requirements increase over time. This means technologies that
initially met the demands of the mass market may eventually exceed the needs of
the market. Furthermore, technologies that initially served only low-end custom-
ers (segment zero) may eventually meet the needs of the mass market and capture
the market share that originally went to the higher-performing technology.
5. Technological change often follows a cyclical pattern. First, a technological
discontinuity causes a period of turbulence and uncertainty, and producers and
consumers explore the different possibilities enabled by the new technology. As

producers and customers begin to converge on a consensus of the desired techno-
logical configuration, a dominant design emerges. The dominant design provides
a stable benchmark for the industry, enabling producers to turn their attention
to increasing production efficiency and incremental product improvements. This
cycle begins again with the next technological discontinuity.
6. The first design based on the initial technological discontinuity rarely becomes
the dominant design. There is usually a period in which firms produce a variety of
competing designs of the technology before one design emerges as dominant.
7. The dominant design rarely embodies the most advanced technological features
available at the time of its emergence. It is instead the bundle of features that best
meets the requirements of the majority of producers and customers.
Discussion
1. What are some reasons that established firms might resist adopting a new
Questions technology?
2. Are well-established firms or new entrants more likely to (a) develop and/or
(b) adopt new technologies? Why?
3. Think of an example of an innovation you have studied at work or school. How
would you characterize it on the dimensions described at the beginning of the
chapter?
4. What are some reasons that both technology improvement and technology
diffusion exhibit s-shape curves?
5. Why do technologies often improve faster than customer requirements? What are
the advantages and disadvantages to a firm of developing a technology beyond the
current state of market needs?
6. In what industries would you expect to see particularly short technology cycles?
In what industries would you expect to see particularly long technology cycles?
What factors might influence the length of technology cycles in an industry?
Suggested Classics
Further Anderson, P., and M. L. Tushman, “Technological discontinuities and dominant
designs,” Administrative Science Quarterly 35 (1990), pp. 604–33.
Reading
Bijker, W. E., T. P. Hughes, and T. J. Pinch, The Social Construction of Technological
Systems (Cambridge, MA: MIT Press, 1987).
Christensen, C. M., The Innovator’s Dilemma: When New Technologies Cause Great
Firms to Fail (Boston: Harvard Business School Publishing, 1997).
Dosi, G., “Technological paradigms and technological trajectories,” Research Policy
11 (1982), pp. 147–60.
Rogers, E., Diffusion of Innovations, 5th ed. (New York: Simon & Schuster Publishing,
2003).
Utterback, J. M., and W. J. Abernathy, “A dynamic model of process and product
innovations,” Omega 3 (1975), pp. 639–56.

Chapter One: The Importance of Technological Innovation

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Chapter One: The Importance of Technological Innovation

Uploaded by

Copyright:

Available Formats

Chapter One

and time-consuming, flexible manufacturing technologies now enable firms to seam-

THE IMPACT OF TECHNOLOGICAL INNOVATION ON SOCIETY

FIGURE 1.1 residual did in fact capture techno-

INNOVATION BY INDUSTRY: THE IMPORTANCE OF STRATEGY

The Innovation Funnel

Research Brief How Long Does New Product

Discovery & Preclinical Clinical Trials Approval

The Strategic Management of Technological Innovation

Part 1: Industry Dynamics of

Chapter 2 Chapter 3 Chapter 4 Chapter 5

Part 2: Formulating Technological

Chapter 7 Chapter 8 Chapter 9

Part 3: Implementing Technological

Chapter 10 Chapter 11 Chapter 12 Chapter 13

In Part Two, we will turn to formulating technological innovation strategy. Chapter

5. While government plays a significant role in innovation, industry provides the

Discussion 1. Why is innovation so important for firms to compete in many industries?

Chapter 3. Concepts and definitions for measuring business innovation

OSLO MANUAL 2018 © OECD/EUROPEAN UNION 2018

3.2. Innovation in the Business enterprise sector

3.2.1. Definition of innovation activities and innovation

OSLO MANUAL 2018 © OECD/EUROPEAN UNION 2018

3.10. As introduced in Chapter 2, a product is a good or service (or combination thereof).

OSLO MANUAL 2018 © OECD/EUROPEAN UNION 2018

3.2.2. Division of innovation effort and responsibilities

3.3. Taxonomies of innovation

3.3.1. Innovation types by object: Product and business process innovations

OSLO MANUAL 2018 © OECD/EUROPEAN UNION 2018

3.25. Product innovations must provide significant improvements to one or more

OSLO MANUAL 2018 © OECD/EUROPEAN UNION 2018

Business process innovation

OSLO MANUAL 2018 © OECD/EUROPEAN UNION 2018

the sale of processes as services. The implementation of business process innovations is

Types of business processes

Short term Details and subcategories

OSLO MANUAL 2018 © OECD/EUROPEAN UNION 2018

OSLO MANUAL 2018 © OECD/EUROPEAN UNION 2018

Fourth edition, 2018 (OM4) compared to third edition, 2005 (OM3)

OM3 OM3 subcomponents OM41 Differences

1. Additional granularity is possible by disaggregating the detailed descriptions in Table 3.1.

OSLO MANUAL 2018 © OECD/EUROPEAN UNION 2018

Combinations of several innovation types by object

Business model innovations

OSLO MANUAL 2018 © OECD/EUROPEAN UNION 2018

3.3.2. Types of innovation according to novelty and impacts

OSLO MANUAL 2018 © OECD/EUROPEAN UNION 2018

3.4. Changes that are not innovations

OSLO MANUAL 2018 © OECD/EUROPEAN UNION 2018

OSLO MANUAL 2018 © OECD/EUROPEAN UNION 2018

3.5. Innovation and business profiling

3.5.1. Innovative and innovation-active firms

OSLO MANUAL 2018 © OECD/EUROPEAN UNION 2018

Table 3.3. Innovative and innovation-active firms

The firm has innovation activities in the observation period

3.6. Use of innovation definitions in data collection

3.6.1. Use of the term innovation in surveys

OSLO MANUAL 2018 © OECD/EUROPEAN UNION 2018

3.6.2. Innovation profiles

3.6.3. Priorities for data collection about innovations

OSLO MANUAL 2018 © OECD/EUROPEAN UNION 2018

Defining and measuring R&D:

FRASCATI MANUAL 2015 © OECD 2015