Professional Documents
Culture Documents
“Trusting your gut on people issues turns out to be a bad idea. Analytics on
your workforce is the most rapidly growing field of analytics. The Power of
People is an excellent guide to this important and burgeoning topic.”
—Thomas H. Davenport, Distinguished Professor, Babson College, and
Research Fellow, MIT Initiative on the Digital Economy
“I believe you will find, like I did, that the frameworks and insights in The
Power of People offer valuable steps toward realizing the potential of your
workforce to create sustainable strategic success.”
—Dr. John Boudreau, Professor, Marshall School of Business; and Research
Director, Center for Effective Organizations, University of Southern
California.
“We are barreling along toward the collision between Big Data, Analytics,
and the successful acquisition, development, and retention of people in our
organizations. The Power of People gives data-led comfort and practical
guidance to business leaders that shows we not only can survive the collision,
we can harness its potential and emerge with a stronger workforce that is
motivated for business and personal success.”
—China Gorman, Board Chair, Universum Americas
“The Power of People is a great book for those who want to build, refine, or
fundamentally improve their HR Analytics offering. The authors have clearly
undertaken some extensive research and are drawing on the experience of a
wide range of people analytics experts. As a result, their book is full of great
advice and can be considered a really good guide for those wanting to realise
the full potential of workforce analytics in their organisation.”
—Dr. Martin Edwards, Kings College London Business School
NIGEL GUENOLE
JONATHAN FERRAR
SHERI FEINZIG
Cisco Press
800 East 96th Street
Indianapolis, Indiana 46240 USA
For information about buying this title in bulk quantities, or for special sales
opportunities (which may include electronic versions; custom cover designs;
and content particular to your business, training goals, marketing focus, or
branding interests), please contact our corporate sales department at
corpsales@pearsoned.com or (800) 382-3419.
All rights reserved. Printed in the United States of America. This publication
is protected by copyright, and permission must be obtained from the
publisher prior to any prohibited reproduction, storage in a retrieval system,
or transmission in any form or by any means, electronic, mechanical,
photocopying, recording, or likewise. For information regarding permissions,
request forms, and the appropriate contacts within the Pearson Education
Global Rights & Permissions Department, please visit
www.pearsoned.com/permissions/.
ISBN-10: 0-13-454600-8
******ebook converter DEMO Watermarks*******
ISBN-13: 978-0-13-454600-1
1 17
Editor-in-Chief
Greg Wiegand
Editor
Kim Boedigheimer
Editorial Assistant
Cindy Teeters
Managing Editor
Sandra Schroder
Cover Designer
Chuti Prasertsith
Production Manager
Dhayanidhi
******ebook converter DEMO Watermarks*******
Copy Editor
Krista Hansing Editorial
Proofreader
Sathish Kumar
Indexer
Erika Millen
Compositor
codeMantra
2 What’s in a Name?
Focus of the Function
Activities of the Function
A Name Fit for the Future
Summary
4 Purposeful Analytics
******ebook converter DEMO Watermarks*******
A Model for Purposeful Analytics
Project Sponsors
Why Do Analytics Projects Fail?
Summary
6 Case Studies
Eight-Step Methodology
Case Study Improving Careers Through Retention Analytics at Nielsen
Case Study From Employee Engagement to Profitability at ISS Group
Case Study Growing Sales Using Workforce Analytics at Rentokil
Initial
Case Study Increasing Value to the Taxpayer at the Metropolitan Police
Case Study Predictive Analytics Improves Employee Well-Being at
Westpac
Summary
II Getting Started
16 Overcome Resistance
Resistance to Workforce Analytics
Stakeholder Skepticism
******ebook converter DEMO Watermarks*******
Financial Frugality
HR Hesitancy
Summary
Glossary
References
Index
I would like to thank my wife, Magdalena, my daughters, Mia and Olivia, and
my parents, Geoff and Aivi.
—Nigel Guenole
I would like to thank my late Great Uncle Bill (William Ferrar), who sent me
two of his own books on calculus as a gift when I was 12 years old and
spurred me on in my own life to write and share my experiences with others.
I thank my son, Arthur, who has brought much joy to my life and reminded
me throughout the writing of this book of the importance of balance in life. I
also wish to thank my parents, James and Janet, and sister, Melloney, who
have supported me through the peaks and troughs of life and for keeping me
grounded.
—Jonathan Ferrar
The human race’s quest for information is never ending. Businesses and
organizations are no exceptions. Business leaders continually seek out
knowledge about their organizations to gain insights from all the data that
exist so that they can make evidence-based decisions to improve the
organization’s performance and gain competitive advantage in the
marketplace.
The discipline called analytics exists to meet this need. Analytics
concerning human resources, people, and the workforce is known as
workforce analytics. The Power of People explores how to establish,
operate, and lead workforce analytics to better serve organizational
ambitions.
All of these resources provide an excellent start for the topic of analytics
******ebook converter DEMO Watermarks*******
applied to work. However, we still felt there was a gap in the market for more
detailed guidance on how a wide variety of organizations can successfully
implement workforce analytics. This topic is the focus of the book you are
now reading.
So how did our book come about? We three authors first met in 2013. We
come from different cultural, national, and professional backgrounds, but
between 2013 and 2015, we collectively and individually wrote several
articles and undertook research on topics related to people, work, and
analytics. Something important happened in spring 2015. Together with a
fourth colleague, we wrote and published a paper called “Starting the
Workforce Analytics Journey: The First 100 Days.” The paper was launched
at an analytics conference in New York, with 50 copies available for a free
takeaway. Early on the first day of the conference, we discovered that all 50
copies of the paper had been taken. We printed another batch, and all of those
also disappeared during the conference. The overwhelming feedback from
conference attendees was that it was the first document people had read that
gave a structured approach and practical tips on how to undertake workforce
analytics.
Within a few days we published an infographic and other material. Then we
took a step back to discuss the success of the paper. Clearly, we had only
scratched the surface with our paper; a book would deliver much more
practical guidance to our thirsty audience. And so the book began. Over the
next several months, we met many people and interviewed scores of experts
in the analytics space—academics, consultants, practitioners, HR leaders, and
data scientists. We cannot thank those people enough for the insights they
provided, which helped shape this book into what it has become.
Our Approach
Building and running a workforce analytics function and delivering
meaningful projects that improve business performance can be complicated,
but learning from the experiences of others can help in successfully
navigating the journey. As we collected ideas from others, we amalgamated
those into four parts in this book.
For Reference
At the end of the book, we provide a glossary that gives standardized terms
and definitions for important elements of workforce analytics. Many
analytics leaders requested this list to aid them in meaningful discussions
with business leaders without getting lost in confused terms and
misunderstandings. Using standardized terms, we can build a professional
common understanding of workforce analytics.
Practical Tips
A growing number of people around the world are involved in the field of
workforce analytics. We were in touch with many of these people while
writing this book. We talked to speakers and attendees at conferences, we
Adoption of Analytics
Many organizations are already realizing the benefits of analytics. A 2014
PWC report, written by The Economist Intelligence Unit, found that 89
percent of large company executives surveyed either already were using Big
Data to make decisions or planned to start doing so in the next three years. In
******ebook converter DEMO Watermarks*******
HR specifically, in its 2016 CHRO report, IBM found that the number of
chief human resources officers (CHROs) using predictive analytics to make
more informed workforce decisions across HR activities had increased by
approximately 40 percent over two years. The evidence of the trend is clear,
as Deloitte’s Global Human Capital Trends 2017 reports: “People Analytics,
a discipline that started as a small technical group that analyzed engagement
and retention, has now gone mainstream.”
A second project, this time at global pest control firm Rentokil Initial,
focused on the predictability of sales success. The project isolated the key
******ebook converter DEMO Watermarks*******
behaviors of high-performing sales professionals and used automated
assessment techniques to select future candidates based on those behaviors.
Global sales rose more than 40 percent and the project had a return on
investment of more than 300 percent.
These examples, described in more detail in Chapter 6, “Case Studies,”
demonstrate that workforce analytics can contribute not just to improving the
effectiveness of HR processes, but also to improving and predicting business
outcomes such as profitability and sales.
Democratization of HR
At a time when data are more readily available than ever, HR is being asked
for more information, better insights, and more precise recommendations to
help executives and managers run their businesses. This puts a strain on the
traditional HR function that primarily dealt with the process side of
recruitment, resourcing, development, and employee relations. For the last 40
years or so, HR has delivered structured programs, developed policies, and
implemented best practices to allow executives and managers to manage
people in a cyclical pattern—for example, through annual performance
reviews, specific salary increase programs, and succession planning cycles.
However, the demand has changed and new requests for information are
emerging, as Table 1.1 illustrates.
Consumerization of HR
Bringing the type of customization experienced by consumers to the world of
work can yield great benefits. In 2012, Amazon reported a 29 percent
increase in second quarter fiscal results. A Fortune article at the time
discussed how Amazon’s recommendation engine contributed much to that
success by using algorithms to heavily customize the browsing experience for
returning customers.
HR can learn a lot from examples like this and begin to use its data to create
predictive models for the “workforce of one” (a term referring to
personalized employee experiences in Accenture’s report “The Future of HR:
A Radically Different Proposition”). But more than this, workers are starting
to expect similar customization from their employers. Many workers would
appreciate recommendations to improve their working experience. This
change is referred to as the consumerization of HR, further discussed by
Mark Feffer in a 2015 Society for Human Resource Management article
focused on recruitment: “Today, job seekers are thought of as customers.”
Examples of workforce personalization include the following:
• Recommendation of modular courses to enhance employees’ skills
• Information on benefits relevant as a worker enters new life stages (for
example, a new baby, marriage, or house purchase)
• Internal job and career moves that best meet a worker’s skills and
expertise
• Opportunities to contribute to projects across the business based on an
individual’s expertise and knowledge
• Provision of performance feedback in real time through manager- to-
employee and peer-to-peer social feedback
“Over the next 10 years, we will see work liberated from the idea of a job.
Work will be disaggregated and re-combined in ways that better suit
employers and employees.”
—John Boudreau, Professor of Management and Organization, and Research
Director of the Center for Effective Organizations, University of Southern
California
It’s a new world of work. Ian Bailie of Cisco summarizes this well: “It’s
about understanding the skills and capabilities of the internal workforce;
dealing with the new gig economy, contractors, and freelancers, and
understanding their skill set; getting better at moving people around the
******ebook converter DEMO Watermarks*******
organization; and enabling them to build careers and their own personal
brands. This will become a dataset that we don’t have today.”
This is an important time for the HR profession to adapt and create
momentum in the field of workforce analytics to capitalize on the changes
shaping the future world of work. As Max Blumberg, founder of Blumberg
Partnerships, Ltd., stresses: “You’d have to be a very brave human resources
director to say you’re not taking analytics seriously.”
Summary
Workforce analytics is a discipline that is increasingly needed in
organizations. This growing demand can be attributed to the following:
• The continued need for increased business value and market
competitiveness
• The requirement for information and data in real time from managers and
executives, to help them run their operations more efficiently and
effectively
• The move toward a consumerized working environment and the provision
of personalized services using workforce-related recommendation engines
• The deconstruction of traditional business models and the proliferation of
the gig economy and independent workers
• The ongoing explosion of Big Data from devices such as wearables and
sensors that will expand the amount of available workforce-related data
1 For pioneering work in this field, see the contributions of Mark Huselid,
Distinguished Professor of Workforce Analytics, at Northeastern
University.
2 In 1999, Shutterfly, Inc., began as a company that helped people print 4-
by-6-inch photographs from their digital cameras. Today it is an industry
leader for photo and video storage, award-winning photo books, gifts,
home decor, premium cards, invitations, stationery, and much more.
Shutterfly is headquartered in Redwood City, California
(www.shutterflyinc.com).
3 Mark Huselid spent more than 20 years at Rutgers University and the last
Summary
Taking into account all of these points, the most descriptive and accurate
name for this function is workforce analytics. This term best describes the
broadest set of workers that contribute to organizational success and the
fullest responsibilities of the function both now and in the future.
Other experts concur with the use of the term workforce analytics as the best
description of the function. Most notably, the SHRM Foundation, the
research arm of the Society for Human Resource Management (the
professional body for the HR profession in the United States), uses the term
workforce analytics in its report “Use of Workforce Analytics for
Competitive Advantage,” undertaken in partnership with the Economist
Intelligence Unit.
With the name of the function in mind, we can define the work of the
function as follows: Workforce analytics is the discovery, interpretation, and
communication of meaningful patterns in workforce-related data to inform
decision making and improve performance.
Clearly, workforce analytics leaders know why they have the role,
what the expectations are, and broadly what they are required to
deliver. A little less clear is the precise job description and the
detailed requirements for successfully performing the role.
This chapter brings clarity to this topic. Of course, the exact skills
needed for the leader of any specific workforce analytics function
vary depending on the size, industry, and geographical complexity of
the organization. Still, most cases share some common requirements.
This chapter provides details on the essential elements that enable a
workforce analytics leader to succeed:
• Internal reporting structure
• Key job responsibilities
• The importance of business acumen and influencing
• Core leadership attributes
Leadership Attributes
Four key leadership attributes are considered most important for the
workforce analytics leader to bring people together as an integrated and
cohesive workforce analytics team:
• Capacity to think
• Willingness to develop others
• Ability to inspire
• Drive to achieve
These attributes ensure that the team achieves more together as a result of the
leader’s actions. They also drive the team members’ engagement with the
business.
Let’s look at each of these four key attributes for success:
• Capacity to think. Many factors will compete for the attention of the
workforce analytics leader: multiple projects, demand for projects (which
can outstrip supply of resources), varied backgrounds of team members,
the need to form a cohesive team, the culture and mind-set of HR
colleagues, and resistance from others. All these leadership demands are
the result of a relatively new team (workforce analytics) in a function
(HR) that is often not considered analytically astute. Such demands
require the leader to be able to think carefully about prioritizing actions.
Furthermore, the projects themselves can be highly complex. They might
involve disparate data sources and complicated analyses to answer
business questions that are, by nature, not easy to answer (if they were
******ebook converter DEMO Watermarks*******
straightforward, they would not likely require the skills of a specialist
analytics team). So in addition to the need to sort through competing
priorities, team leaders must be able to think about very complex projects.
• Willingness to develop others. Team members who are perfectly
equipped to perform workforce analytics roles are in short supply. Back in
2011, McKinsey & Company reported that the United States would
experience a shortage of at least 140,000 people with deep analytical
skills by 2018. Certain skill sets likely will need to be hired and
developed; this is particularly true for data science, the most demanded
job in the United States in 2017, according to CareerCast. Although hiring
and developing skills for any team is a typical role of all leaders, it is
particularly important in workforce analytics for the following reasons:
• The profession of workforce analytics is very young. As such, more
commitment is needed initially to develop people, especially given the
short supply of certain skills.
• The team will come from varied backgrounds, so more cross-training in
a variety of skills such as statistics or financial literacy will be needed.
• The team will need to learn new and unfamiliar skills, such as
consulting and data science.
• All workforce analytics professionals who deal with internal clients
must be proficient in stakeholder management, another skill that can be
difficult to teach and learn.
Leaders need to integrate the work of people from different backgrounds
and disciplines into a cohesive team. They also need to nurture and grow
talent, take a developmental outlook toward employees, and show a
passion and drive for collective action and teamwork.
• Ability to inspire. Workforce analytics leaders need to instill a belief in
the team to succeed, often in the face of limited resources and external
pressures. Effective leaders do this by influencing key people and using
others to exert influence when necessary. They build confidence in their
team members’ ability to succeed and inspire them to work as a coherent
team around a common vision and mission. This ability is important
because the team might encounter skepticism and resistance from other
HR professionals and stakeholders (see Chapter 16, “Overcome
Resistance”).
******ebook converter DEMO Watermarks*******
• Drive to achieve. Workforce analytics leaders need tenacity and
resilience. Analytics projects can be complex in both the business
questions they seek to answer and the analytical methodology needed to
derive insights and recommendations. And projects don’t end there (see
Chapter 4, “Purposeful Analytics”): Workforce analytics projects also
require implementation to drive change in the business. All this means
that projects can take weeks or months to undertake; in some cases,
delivering a return on investment (ROI) can take years.
To ensure that the workforce analytics work gets done, team leaders must
have a proactive nature, a mentality for continuous improvement, and a
strong customer focus mindset.
BE EGO-LESS
Alexis Fink is the General Manager of Talent Intelligence &
Analytics at Intel1 and has enjoyed success in business analytics
across several organizations. She has one main piece of advice for
workforce analytics leaders: “Don’t make it about yourself.”
She explains that some people forget that analytics is about the
business and become consumed with their own self-importance.
“Their ego gets in the way and analytics projects begin to succeed or
fail due to the person behind the analytics. This is wrong.”
Alexis suggests three strategies to stay grounded:
• Don’t try to be a “know-it-all.” As an analytics leader, you have
access to all the data, but you don’t necessarily know best. Help your
leaders and colleagues understand the workforce-related data and
lead them on a journey. As Alexis recommends, “Be nonthreatening.”
• Keep pushing forward. Get to know your leaders. Deliver what
leaders ask for, and then introduce additional insights so they get
more than they asked for. This way, you will build your credibility.
As Alexis points out, “You have to serve the appetizer to get to the
main course.”
• Don’t get yourself excluded from the big projects. Learn the business
and understand who the decision makers are. Find a way to bring
analytical insights into the fold. Then impress them with your team’s
******ebook converter DEMO Watermarks*******
work.
Alexis summarizes, “I learned to be ego-less. I had to get really good
at getting my ideas to come out of someone else’s mouth.”
Bringing the team together around a common goal and leading them to
achieve great things is as important for workforce analytics leaders as it is for
other leaders in the business. Encouragingly, scientific research suggests that
the vast majority of leadership capabilities are learned from experiences, and
the best experiences are often on the job. Morgan McCall, a professor at the
University of Southern California, sees the following formula for acquiring
leadership skills: 70 percent of the learning should be acquired on the job, 20
percent of the learning should come from other people (for example,
experiences with good and bad bosses), and the remaining 10 percent should
come from classroom-based learning such as formal leadership training
programs.
Beyond the leadership attributes described here, workforce analytics leaders
also need a credible level of familiarity with each of the team’s
specializations. Although leaders are not likely to have the same
specialization depth as individual team members, they should have, for
example, a working knowledge of statistics and a good feel for numerical
information. For more on the core skill areas for the workforce analytics
team, see Chapter 12, specifically the section “Six Skills for Success.”
Summary
The leader of the workforce analytics team need not have followed a specific
career path before taking on the leadership role, but certain approaches,
attributes, skills, and experiences will provide a better platform for success:
• Have the workforce analytics leader report directly to the CHRO.
• Clarify the role of the workforce analytics leader and the specific
responsibilities for success.
• Develop business acumen by continuously learning about the internal
operations, metrics, and key performance indicators for the business.
• Build external awareness by learning about the marketplace, competitors,
and other external factors.
Project Sponsors
Throughout this chapter, you have seen project sponsors identified as a
critical party. In this section, we discuss the relationship between the
workforce analytics practitioner and the sponsor and explore why
******ebook converter DEMO Watermarks*******
sponsorship is important for success. We also map the analytics practitioner’s
requests of the project sponsor to each step in the purposeful analytics
methodology.
Project sponsors are individuals who have an active interest in one or more
specific workforce analytics project. They are usually responsible for
approving the project, providing resources needed for project execution,
advocating for the project, and working with stakeholders to ensure that
actions are implemented. In short, the sponsor has a vested interest in the
project from start to finish.
How do you know whether you have the right sponsor to help you deliver a
successful analytics project? Strong sponsors are highly respected leaders in
their organizations, with a deep understanding of the organization and the
challenges it is facing. They are well connected and able to rally support, and
are willing and able to secure the resources needed. Finally, the best sponsors
are highly motivated and available to see the project through to completion
and benefit realization. Damien Dellala, Head of People Data & Analytics
Enablement at Westpac Group, emphasized the importance of a good
sponsor: “Experience has taught me that, when starting a workforce analytics
project, having an engaged sponsor is at the top of the list of success factors.”
When securing sponsorship, setting expectations on outcomes before
undertaking an analytics project is important. This means thinking through
the possible scenarios and posing them to the sponsor. For example, if the
issue is high turnover among a certain group of employees, and the analysis
reveals compensation is a strong contributing factor, is the sponsor willing to
allocate budget to solve the problem? What if lack of career growth is a root
cause—is the organization willing to design new career guidance and
promote people at a more rapid pace than it has traditionally done? Thinking
through these possible scenarios and building commitment and willingness to
act increases the probability that the organization will actually implement the
necessary actions.
Figure 4.3 summarizes the many actions required of the project sponsor to
ensure analytics success.
Project Outcomes
Projects can fail when results are poorly communicated, no action results, or
action occurs but with no evaluation to determine the business return. These
factors can hinder success:
• Poor visualization with unclear insights
• Ineffective storytelling that loses the simplicity of the message or fails to
articulate it
• No buy-in for the outcomes of the analytics project, with no accepted or
implemented recommendations
• No clear plan or owner to implement the organizational change
recommended by the project
• Lack of evaluation to determine the project’s impact
Summary
Discovering, interpreting and communicating meaningful patterns in
workforce-related data demands a robust methodology. Whether for a short
project of a few weeks or a longer one that spans many months, methods
need to be straightforward and complete. Successful workforce analytics
projects follow these main methodology recommendations:
• Consistently apply all eight steps of the purposeful analytics model to
your analytics projects.
• Start projects by defining the business question and building strong
hypotheses.
• Use the most appropriate technology and methods for your analyses to
ensure robust results.
• Get the point of your analytics projects across to your sponsors and
stakeholders through clear visualization and storytelling.
• Help your business translate project recommendations and decisions into
action and evaluate the results.
• Choose your sponsors wisely, with an eye toward strong, consistent
support throughout the entire lifecycle of your projects, from request to
implementation.
• Familiarize yourself with the common reasons analytics projects fail, and
throughout the eight steps, ensure that potential problems are avoided or
addressed.
Research Design
Before quantitative or qualitative analyses can occur, decisions must be made
regarding what data will be collected, when it will be collected, how it will be
******ebook converter DEMO Watermarks*******
collected, and from whom it will be collected. These questions fall under the
topic of research design. The research design you apply determines how
rigorously you can reach conclusions about causes and effects following your
analysis. Identifying cause and effect is not the only goal in workforce
analytics, but it is a common one. You are unlikely to unequivocally show
causal effects from single research studies. However, designs that allow more
confidence in causality increase the probability of creating successful
interventions.
Research designs can be categorized by an effectiveness hierarchy,
summarized in Table 5.1. The strongest research designs are randomized
experiments, followed by quasi-experimental designs, observational or
correlational designs, and, finally, qualitative designs.
Without a thoughtful research design, even the most sophisticated analyses
will likely prove fruitless. Work with your data scientists to select a strong
research design before you begin your analyses.
Experimental Designs
In a randomized experiment, you generally have two groups: one that
receives an intervention (the experimental group) and one that does not (the
control group). Employees are randomly allocated to either group. The
randomization means that the groups are, on average, equivalent in every way
at the start of the experiment.
After an intervention with the experimental group, the groups are measured
******ebook converter DEMO Watermarks*******
on the outcome under study. Consider, for example, a project that examines
the effect on engagement after giving employees more autonomy. In an
experimental approach, any difference in engagement between the groups
must have been caused by the intervention, assuming there are no
confounding factors (such as groups becoming aware of the goals of the
experiment). This is because before the experiment, the groups had
equivalent engagement due to randomization. In addition, the control group
did not receive the intervention that increased autonomy. A key point to note
is that variables under study are manipulated.
Randomized experiments, therefore, meet the three criteria for showing a
causal effect:
• The cause happens before the effect.
• The experiments show a relationship between the cause and the effect
(when a change occurs).
• Other possible causes are ruled out due to randomization.
The design can be further strengthened by statistically controlling for a range
of other plausible causal variables. This can help rule out alternative
explanations of findings, in case the randomization was not perfect. For
example, continuing with the autonomy and engagement example, if the
randomization did not lead to equivalent personality types across groups, the
results could be jeopardized because different personality profiles might
cause different preferences for autonomy across groups. In these cases,
worker personality could be measured and the results of the analyses adjusted
for any differences between the groups.
Quasi-Experimental Designs
Conducting randomized experiments in everyday working life is often not
possible. We cannot always isolate the variables we want to study, and
randomly allocating workers to different conditions is often impossible. As a
result, other designs are chosen that still allow some confidence in identifying
a causal relationship, even though they are not as strong as in a randomized
experiment.
The quasi-experimental design is similar to the randomized experiment, aside
from one key feature: The employees are not randomly allocated to the
experimental and control conditions. This limits the strength of inferences
******ebook converter DEMO Watermarks*******
that can be drawn from analyses when compared to randomized experimental
designs because the lack of randomization makes it impossible to confidently
declare that the two groups were equal, on average, before the experiment.
Despite these limitations, robust conclusions about effective interventions can
still be drawn from quasi-experimental designs in workforce analytics,
particularly if results replicate across multiple studies. A quasi-experimental
approach to the autonomy and engagement topic might involve delivering the
intervention to two different business units because randomization is often
unfeasible; treating two people differently is difficult if they are working in
the same team or unit. With these designs, it is not possible to account for
other factors, such as different managers giving different levels of
recognition; those factors might indeed have caused differences in
engagement levels. Nevertheless, this design often provides enough
confidence to decide whether to continue the intervention.
Correlational Designs
Designs that do not involve randomization and manipulation, or a control
group, are referred to as correlational or observational. In these studies,
analytics professionals observe the way variables relate to one another
without inferring a causal connection between variables. Strong dependable
correlations (for example, sizable correlations based on large random
samples) can still be useful in workforce analytics, but it is important to
understand when a causal association cannot be confirmed.
Whereas the experimental and quasi-experimental approaches to the
autonomy and engagement example included a control group, correlational
designs typically have no control group. Instead, the existing levels of
autonomy and engagement are assessed for all individuals, and the strength
of the association is studied. Inferring causal associations from these designs
is difficult, but they may be useful in identifying variables for further study
using experimental or quasi-experimental designs. These designs lead to
stronger conclusions than when relying on intuition. Advanced techniques
can help identify causal relationships by analyzing data from correlational
designs (for example, propensity scores and instrumental variable
techniques), although their effectiveness varies depending on the situation.
Qualitative Studies
******ebook converter DEMO Watermarks*******
In contrast to the designs discussed so far, which all involve numerical
assessments of the relationships between variables, the hallmark of
qualitative studies is that they try to understand organizational phenomena
from the perspective of workers without using quantitative methods.
Examples include ethnographic studies and focus groups. A qualitative study
that examines the association between autonomy and engagement might
involve interviews or focus groups with workers to discuss their experiences
of autonomy and engagement. An ethnographic approach might have the
researcher work for a period of time in the job, side by side with actual
workers, to understand how autonomy relates to engagement through the eyes
of the workers. It is possible to use qualitative and quantitative methods to
complement one another, a topic we discuss in the section, “Qualitative
Analysis,” later in this chapter.
On their own, qualitative methodologies do not provide the level of
confidence regarding correlational or causal effects that many business
leaders and analysts want to see before they act on recommendations from
workforce analytics. For this reason, we place qualitative studies at the lowest
level of the hierarchy of research designs for causal inference (see Table 5.1).
Objectives of Analysis
The overview of analysis objectives (quantitative and qualitative) presented
in the following sections and in Figure 5.1 represents the level of detail that
Quantitative Analysis
In workforce analytics, most quantitative analysis (that is, statistical analysis
of numerical data) aims to do one of the following: explore, associate,
predict, classify, reduce, or segment information about employees and
organizations.
• Explore. Exploratory analysis helps understand your variables. Analysis
might involve summarizing the data with a statistic such as the average, or
looking at how spread out the values of the variable are, using a statistic
known as the variance (that is, the variability) of the variable. Exploring
might involve identifying extreme cases or determining the extent of
missing data. Analysis focused on exploration often uses simple graphing
******ebook converter DEMO Watermarks*******
techniques to reveal the distribution of the data (for more on these topics,
see Chapter 10, “Know Your Data”). More advanced exploratory analysis
might check differences in scores on variables for known groups, such as
men versus women, or ethnic minority versus majority groups.
Exploratory analysis is particularly useful for monitoring and reporting
demographic trends, as well as for preparing data for more advanced
analyses. Approaches include techniques such as t-tests or analysis of
variance, referred to as ANOVA. Exploring data involves understanding
your data, preparing your data for more advanced analysis, and testing
simple hypotheses.
• Associate. One goal of analytics is to look at the relationship, or
association, between variables that can take on any value between their
minimum and maximum possible values (for example, age or tenure),
with a higher score indicating more of the variable. An example might be
the relationship between extroversion and sales performance. Sales
performance can likely take on any value within a plausible minimum and
maximum, so it is considered a continuous variable. Rating scales
common in surveys (for example, strongly disagree to strongly agree) are
often analyzed as continuous.
Relationships between these types of variables are usually studied using
methods that estimate correlations. The most common correlation,
Pearson’s product-moment correlation, has a possible range from –1 to
+1. A value of –1 means that two variables are perfectly negatively related
(that is, as one increases, the other decreases by an equivalent amount). A
value of 0 indicates no relationship, and a value of +1 means the variables
are perfectly positively related (that is, as one increases, the other
increases by an equivalent amount).
• Predict. When analyses are used to make forecasts about the future, such
as estimating the value of a continuous variable of interest (referred to as
an outcome variable or dependent variable), the analyses are referred to as
predictions.
Given what we currently know about employees, the general pattern in
predictive analysis is to estimate their future behavior at work. For
example, if we know that the relationship between levels of extroversion
and sales performance is strong, sales performance should improve if we
hire people who score high on an extroversion assessment. Methods used
******ebook converter DEMO Watermarks*******
for making predictions include linear and multiple regression, regression
trees, neural nets, support vector machines, and time series analysis.
• Classify. Classification analyses can be thought of as analogous to
association, but this analysis focuses on associations when the outcome
variable is discrete. A discrete variable has a limited number of possible
categories and does not have an intrinsic ordering. It is sometimes called a
nominal variable. An example of a discrete variable might be whether a
worker receives a promotion or whether an employee leaves the business.
In such cases, the outcome is referred to as a discrete or categorical
outcome (not a continuous outcome).
Methods such as correlation and regression have been adapted to deal
with these types of variables. Instead of finding correlations, chi-squared
tests examine associations for categorical variables; instead of using
regressions to make predictions, techniques such as logistic regression are
used to make classifications. For example, the risk of an employee leaving
(which has only two possible values, stay or leave) can be expressed as a
probability; employees with a risk of leaving can be looked at more
closely for some sort of intervention.
• Reduce. Workforce analytics teams analyze data sets that often contain
hundreds or thousands of variables. This is too much information to make
sense of without combining some variables. The primary goal of some
statistical techniques, such as principal components analysis (PCA) and
factor analysis (FA), is to summarize (reduce) the information in many
different variables to create a smaller number of variables. Having fewer
variables makes analysis and interpretation more manageable.
The basic principle in reduction analyses is to aggregate the similar
variables into fewer variables. Say, for instance, that you have three
variables that measure employee performance: a manager performance
rating, data on whether the employee met critical milestones, and the
success that the employee demonstrated in training. Which of these
variables should you focus on predicting? Statistical analyses such as
PCA and FA can create, if appropriate, a single performance variable out
of other similar performance variables for prediction. This is usually
appropriate only if you can show that all three together are strong
measures of performance, which PCA and FA check. The new variable(s)
can then be explored or predicted without losing too much information.
******ebook converter DEMO Watermarks*******
• Segment. Although the focus of reduction analyses is to group a large
number of variables into a smaller number for exploration or further
analysis, the goal with segmentation is to group the number of cases (for
example, workers) in your data set into a smaller, more manageable
number that provides a meaningful representation of groups in your data.
Techniques in the segment category fall under the general label of
clustering. Cluster analysis puts similar cases into groups, for either
exploration or further statistical analysis. An example application might
involve using cluster analysis to identify subgroups of employees who
have similar responses to a variety of survey questions indicating high
stress so that some form of intervention can be applied to improve their
ability to manage stress.
Qualitative Analysis
The goals of qualitative research are to understand organizational phenomena
without using quantitative data. Techniques include ethnography, focus
groups, and detailed case studies. Objectives of qualitative research in
workforce analytics include hypothesis formulation, interpretation, and
******ebook converter DEMO Watermarks*******
contextualization.
• Hypothesize. Qualitative analyses can be particularly valuable in
generating hypotheses when no theories or quantitative data exist. The
resulting hypotheses can then be tested using quantitative methods. For
example, interpretation and discussion about the reasons people leave a
firm can help in creating a hypothesis that can be tested with quantitative
data. In this example, a qualitative analysis of exit interview data might
suggest that the physical office environment was a factor in decisions of
employees who quit. A quantitative study might test the hypothesis that
employees are more likely to stay if they work in buildings that have been
recently refurbished.
• Interpret. Another objective of qualitative research is to help interpret
quantitative results. Qualitative studies explain why events occur instead
of simply describing the strength of relationships in quantitative terms.
For example, if quantitative data reveal that new hires are taking too long
to get up to speed, qualitative research might explain why that is the case.
In this example, focus groups might reveal that managers are not meeting
with new hires in the first week.
• Contextualize. Michael Pratt, a professor at Boston College, notes that
qualitative approaches are good for contextualizing quantitative results.
Contextualizing aims to explain technical quantitative findings to HR and
business leaders by adding color and depth to illustrate a point. For
example, including verbatim quotes from interviewees can bring to life
the results of quantitative analyses. The explanations can then help turn
the analytics project into a story, emphasizing the perspective of those
impacted by the recommended actions. For an excellent introduction to
contextualization in qualitative research, see Chapter 3, “Qualitative
Research Strategies in Industrial and Organizational Psychology,” by
Tomas Lee and colleagues in the Handbook of Industrial and
Organizational Psychology (American Psychological Association, 2010).
Unstructured Data
Increasingly, data that workforce analytics professionals encounter will be
unstructured, or difficult to store in rows and columns of a “flat” data file.
Examples include video, audio, and the vast amount of text and language that
is available via the Internet.
******ebook converter DEMO Watermarks*******
In the past, much of this information was used qualitatively, to help form
hypotheses, interpret findings, and contextualize results for audiences. Today
these data sets are commonly analyzed quantitatively. Perhaps most advanced
is text-mining of data sets to assess the sentiment of text passages, such as
comments in open-ended surveys. Sadat Shami, Director of the Center for
Engagement & Social Analytics at IBM, describes the potential of text
analytics. “Developments in managing and analyzing unstructured data, such
as text from social media, are really showing the value of social media as a
data source for signals that we can use to infer, for example, the level of an
employee’s engagement.”
Methods for quantitatively analyzing qualitative and unstructured data
generally focus on converting data that were captured as language or images
into a numerical representation. After that point, the objectives of analytical
techniques are the same as those already mentioned for quantitative analysis.
Examples of these techniques are rare event detection, sentiment analysis,
and trending topics.
Summary
Analytically curious HR professionals, and certainly all members of the
Eight-Step Methodology
******ebook converter DEMO Watermarks*******
Our eight-step methodology for workforce analytics is described in detail in
Chapter 4, “Purposeful Analytics.”
The first steps focus on understanding why an analytics project has been
initiated:
Step 1: Frame Business Questions
Step 2: Build Hypotheses
The next steps describe how the project will be conducted:
Step 3: Gather Data
Step 4: Conduct Analyses
Step 5: Reveal Insights
Step 6: Determine Recommendations
The final steps ensure that action will be taken as a result of the project:
Step 7: Get Your Point Across
Step 8: Implement and Evaluate
Understanding the extent and size of the problem and identifying a business
leader to sponsor and benefit from the project allowed the people analytics
team to clarify the intent of the project and frame the business questions (Step
1).
Piyush was confident that his team could define, measure, and understand the
factors causing attrition in a clear, predictable, and sustainable way. He was
******ebook converter DEMO Watermarks*******
also convinced that his team could find insights and make actionable
recommendations to address the problem. When reviewing the attrition
problem, the team identified two specific groups of people they suspected
were a high retention risk, and they tested this suspicion in the following
hypotheses:
Hypothesis 1: Women and diverse employees have higher attrition risk
than men.
Hypothesis 2: Employees who work remotely (for example, at a client’s
location) have higher attrition risk than employees who work from a
Nielsen office.
By isolating one sizeable yet discrete business unit, Piyush and his team
could focus on specific hypotheses to address the business questions (Step 2).
The study was limited to people in the U.S. business, to avoid specific works
council and data privacy challenges. Piyush explains: “We involved the
general counsel and the chief privacy officer and their teams from the start. In
doing so, they helped us by endorsing our project and recommending that we
begin our project in the U.S. only to simplify the gathering of data.”
The team focused on a time period of five and a half years; this was the
longest time period of consistent data, and it gave them sufficient data to
complete a strong analysis.
The two predominant technology systems that the team used to collect the
required data were the SAP Human Resources Information System (HRIS)
and the Oracle Taleo recruitment applicant tracking system.
Choosing and gathering the right data is important for any analytics project,
but Piyush outlines another caution: “Sometimes we wait for data to become
perfect. I believe in ‘design thinking,’ where we imagine the future state but
start building with what we have and keep improving as we go along.”
The next step was analysis. As Figure 6.1 shows, the team decided to use a
Cox regression analysis, an example of the quantitative analysis objective of
classification (see Chapter 5, “Basics of Data Analysis”). Although the team
considered other techniques, such as logistic regression, it chose Cox
regression because that method allowed them to study attrition over time as a
function of the various predictor variables.
In its analysis, the team found no support for the first hypothesis: Women and
diverse employees have no higher attrition risk. But the team did discover
three core factors contributing to attrition: lack of lateral moves,1 being
located at a client site, and recent hire date (tenure of less than one year). The
second factor supported the second hypothesis: Attrition risk was indeed
higher for associates working remotely from a Nielsen location.
These insights were clarified with a high degree of confidence. For example,
someone given a lateral move was proven to be 48 percent less likely to leave
than someone who was not given a lateral move. Interestingly, Nielsen had a
very low percentage of managers offering lateral moves or associates asking
for them; less than two percent of people received a lateral move globally.
Although lateral moves might bring a small compensation change, the new
role essentially has a similar responsibility level as before, but in a different
environment (such as for a different business, manager, or function).
The insights were revealed (Step 5) with confidence due to the strong
quantitative analysis used.
One of Piyush’s guiding principles is that business leaders will get more
excited if financial benefit for analytics can be proven. His next step was to
******ebook converter DEMO Watermarks*******
demonstrate exactly that. Piyush and his team, together with compensation,
talent acquisition, and other HR experts, plus people from the financial
planning and analysis group, built a cost of attrition model based on actual
voluntary attrition data. This model included factors such as lost productivity
and time (and associated cost) to recruit. Working with members of the
finance department, they ensured that any model created would be taken
seriously and be financially validated.
The financial impact analyses revealed that, for every 1 percentage point
decrease in attrition, Nielsen avoided approximately $5 million in business
costs. This analysis got the attention of the senior leaders, as Piyush
elaborates: “We shared the model with our CEO and CHRO—they loved it.
They really liked how we were using analytics empirically and financially.”
With the insights (the factors shown to contribute to attrition) gathered and
the cost of attrition model developed and validated by finance, the team could
focus on building very specific recommendations focused on talent reviews,
lateral moves, and onboarding.
The first major recommendation was to focus on lateral moves as part of
talent reviews. As part of discussions about talent and succession, leaders
were expected to spotlight key individuals who would benefit from lateral
movement to another part of Nielsen. This was embedded in the talent review
process so successfully that it became part of discussions with the CEO
beginning in late 2015.
The second recommendation concerned a program called Ready to Rotate
that already existed but was poorly used. It was designed for associates to
self-identify when they would like to be considered for a lateral move. It was
originally created but not implemented because it lacked a committed
sponsor. With the level of support Piyush had created, he was confident that a
program like this could be reignited and implemented.
Finally, beyond programs supporting lateral moves, recommendations were
implemented around onboarding. These were designed to address one of the
other insights derived regarding the relatively high attrition among employees
with less than one year of tenure. The onboarding actions included a buddy
program and an “information all in one place” system for new recruits to
allow them to more quickly feel connected and integrated.
The team continues to work with business leaders to measure attrition and
advocate the Ready to Rotate program (Step 8).
The analytics team ensured that ISS had clarity on the business priority of
employee engagement to determine a return on the investment (Step 1).
To ensure collective agreement on the aims of the analytics work and strong
management, a project team was established. The team included
representatives from Group HR, Group Marketing, and an external
consultancy. Of particular note was the external partner selected to assist the
team: Morten Kamp Andersen is an experienced business consultant in the
field of analytics and had already worked with ISS and several of its senior
leaders on earlier projects. Simon explains, “It was critical that the right
people, including the main stakeholders, were involved in this project from
the outset. In many projects, collaboration across the business only begins
when the results are presented, but we wanted to establish that collaborative
approach from the start.” With such an approach the team could more easily
get buy-in, support, and the resources it needed to deliver.
One of the project team’s first tasks was to clarify its hypotheses. Two clearly
stated hypotheses captured the expectations about the project and guided
future data collection and analyses:
Hypothesis 1: Employee engagement is positively related to both
employee and customer experience.
ISS identified data sources that were strong indicators of each variable
relevant for the hypotheses (Step 3) and used a variety of statistical
procedures to simplify the dataset and test these hypotheses (Step 4).
The analyses provided support for the first hypothesis at ISS: Employee
engagement was indeed positively related to customer satisfaction. However,
as Simon says: “While this finding reflected the existing external research
suggesting that engagement is associated with performance outcomes such as
customer satisfaction and profitability, the project team wanted to know
more.” The real insight was revealed when the analytics team considered
exactly what aspects of employee engagement were related to customer
satisfaction as measured by the cNPS.
Three aspects of engagement turned out to be particularly strongly linked
with customer satisfaction: motivation, capability, and purpose. In other
words, cNPS scores were higher for business units in which employees were
more motivated to do a good job, were well trained, and understood customer
expectations. Moreover, as the second hypothesis predicted, eNPS and cNPS
were positively related with contract profitability. Figure 6.2 illustrates the
contract profitability (shown as a percentage) as a function of eNPS and
cNPS. It shows that, as both eNPS and cNPS increase, so does contract
profitability.
The analytics team identified motivation, capability, and purpose as the key
drivers of employee engagement, and these had the highest influence on the
cNPS. The next step was deciding what to do about it. The team concluded
that the findings about employee engagement had important implications for
HR processes and strategic initiatives. As a result, the team had the following
recommendations:
• Functional training programs to address the capability factor (for instance,
skills training for facilities cleaners).
• A behavioral training program called Service with a Human Touch, to
******ebook converter DEMO Watermarks*******
focus on understanding the emotional connection between workers and
clients and delivering superior user service. The training was implemented
for first-line managers responsible for contract delivery initially in
Denmark before it was implemented globally.
• A manager education program to address the purpose factor. This would
focus on ensuring that staff knew what both ISS and customers expected
of them.
• A motivation toolkit to address the motivation factor. This would be part
of an existing manager development program.
In addition, ISS was advised to conduct a managerial training needs
assessment to determine where additional training was needed before
delivering the training.
Not content to stop there, ISS wanted an even stronger research design to
increase confidence in its conclusions. As a result, ISS implemented the
recommendations intended to increase engagement among its staff at one of
its customer locations, a financial services firm. This would enable the team
to study the effects of the changes using a pre-/post-research design.
In the study, customer satisfaction was measured with a user survey both
before and after interventions intended to increase the three drivers of
engagement: motivation, capability, and purpose. The interventions included
extensive training for all ISS managers, supervisors, and front-liners on how
to manage and deliver a service against predefined behavioral standards. In
addition, customers were asked to clearly communicate with front-line staff
what they expected in terms of service quality and standards. Evaluation of
this follow-up study revealed a significant increase in employee engagement
following the training, as well as a significant increase in customer
satisfaction.
Encouraged by the findings from this analytics project and their ability to link
engagement to business outcomes, ISS team members are now exploring
links between engagement and sickness rates, as well as customer churn. ISS
says that what made this project successful was the team’s ability to link HR
practices (notably, efforts to increase engagement) with business outcomes
outside of HR (customer satisfaction and, ultimately, contract profitability).
This analytics project had strong sponsorship from the CEO and a clear
business problem (Step 1). Multiple hypotheses existed and needed to be
validated and verified (Step 2) as part of the analytical methodology.
The iterative process of collecting data (Step 3), analyzing it (Step 4), and
discovering insights (Step 5) led to the recommendation to implement a
single selection test (Step 6). Strong statistical evidence backed up the
recommendation, and the financial impact was quantified.
Overall, this project demonstrated a clear and direct business impact in terms
of increased sales. It succeeded because of its high-level sponsorship,
effective stakeholder management, and strong methodical approach to
analytics. As Steve summarizes: “This project demonstrated how analytics
can shape the future through helping people secure the right jobs that will
make them successful and bring benefit to business leaders and owners
through increasing sales. It’s a win–win for everyone.”
Martin and his team found that spending time framing the business question
(Step 1) and clarifying the hypotheses (Step 2) was important in providing
focus to the analytics team.
The phased approach helped the HR team gain credibility, establish metrics,
and gather the data required to undertake more complicated analyses (Step 3).
With consistency and clarity around the Met’s current data established,
Martin and his team set out to visualize what the future workforce could look
like. They analyzed potential scenarios using “what if” models. Using Excel,
they built an analytical model embedded with forecasting calculations to
provide insights on minorities, gender, recruitment, and several more aspects
of the entire workforce. Martin confirms, “This visual model was greatly
appreciated and game changing for the Met. It was updated and distributed
monthly to all stakeholders.”
Focusing back on the core topic of diversity and workforce costs, Martin was
able to apply more analytical models to the work. For example, the team
forecast attrition for police officers using historical data, predicted
recruitment targets, and forecast year-end headcount.
The analytics methods gave the HR team at the Met significant credibility.
Clare Davies had previously described a “we can’t rely on HR data”
sentiment that had prevailed across the organization before this project. After
just a few months, and with the implementation of the analytical models and
methods, the mood changed. Clare explains: “We created a relentless focus
on accurate data and insight, regularly using it to improve aspects of the
operations. We worked hard to ensure that finance and HR data reconciled so
that we had one version of the truth. We involved stakeholders along the way
and gained credibility.”
Various models were used to manage diversity, hiring, and workforce costs
(Step 4). These allowed the workforce analytics team within HR to gain
credibility.
Conducting accurate analyses and sharing the resulting data and insights was
Identifying the right data sources (Step 3) and their choice of analyses (Step
4) allowed the team to appropriately test the hypotheses, lending confidence
to the outcome. Interpreting the results of the statistical models allowed the
team to derive insights about the specific drivers of stress (Step 5).
The team made recommendations to put findings into action (Step 6), and got
its point across (Step 7) to key stakeholders, energizing and encouraging
them to act. Team members worked with the business to operationalize the
recommendations by delivering insights to managers where and when
needed; they also established a plan to follow through, to ensure that the
expected value would be delivered (Step 8).
This methodical approach allowed the HR Analytics team, together with the
Health, Safety, and Wellbeing team, to effectively address Westpac’s
challenge and create an environment that helps its people flourish and grow.
Damien says he and his team enjoyed being part of this feel-good project
because they “helped solve a business problem to help employees be their
best selves at work.”
Summary
The expert analytics professionals in these case studies undertook their
projects with focus and intent. The organizational challenges and the
techniques used to address them varied across the cases, but the following
factors were common to all:
• The problem to be addressed was clearly articulated and linked to the
******ebook converter DEMO Watermarks*******
overall strategy of the organization.
• Sponsorship of the project by an influential person positively impacted all
stages, from initiation through implementation.
• Hypotheses were clearly understood and testable (and, in some cases,
developed iteratively).
• Data gathered and analyses undertaken were appropriate for the project.
• Insights and recommendations were clear and precise.
• Communications to various stakeholders were well planned and
developed with the help of expert communications professionals, when
needed.
• Projects were designed to provide business impact and were evaluated for
success and learning.
• Internal or external partners contributed expertise to the team to ensure
success.
“You need to know why analytics is required in the first place, whether that’s
because there’s ‘smoke in the buildings’ or because the scale of the
enterprise means managing the workforce numerically is the only option.”
—Ian O’Keefe, Managing Director,
Head of Global Workforce Analytics, JPMorgan Chase & Co.
“I choose my projects carefully and make sure I understand them. You have
to think it all the way through. Strong analysis is an accepted prerequisite,
but you need to know what are you going to do with the results and how you
will implement the recommendations. You need to think through the project
all the way to the individual workers who will be impacted. If you don't do
this, you will find it difficult to demonstrate why your project will make a
difference to the business and, therefore, it will be difficult to get funding.”
—Simon Svegaard
Business Analytics Manager, ISS Facilities Services
No magic formula covers writing the ideal vision and mission statements.
However, the statements should be well informed by the early views of your
key sponsors about why they believe a workforce analytics function is needed
and what the demand is for workforce analytics projects. If you have to dig
out documents every time you need to refer to them, your vision and mission
statements are not serving their purpose.
Summary
These key steps set the direction for workforce analytics in your organization:
• Pause, listen, and think before you dive into your first analytics project.
• Talk to the person who appointed you, or the most senior person in your
functional line (for example, the CHRO), to clarify the scope of your role.
• Identify prospective project sponsors for interviewing and ask them about
business challenges.
• Identify which of the Seven Forces of Demand are driving the primary
need for analytics in your organization.
• In a memorable vision statement, describe the desired future impact of
your function on the organization.
• In a memorable mission statement, describe your objectives and how you
will address them.
• Use the vision and mission statements to communicate your function’s
identity to your team and the wider organization, and to prioritize your
1 BNY Mellon is the corporate brand of the Bank of New York Mellon
Corporation. Its heritage dates back to 1784. As of December 31, 2016,
BNY Mellon had $29.9 trillion in assets under custody and/or
administration, and $1.6 trillion in assets under management
(www.bnymellon.com).
2 The Seven Forces of Demand is a copyright of the authors of this book:
Nigel Guenole, Jonathan Ferrar and Sheri Feinzig.
3 Agoda.com is one of the world’s fastest-growing online hotel platforms.
Established in 2005 as a start-up, Agoda.com expanded quickly in Asia
and was soon acquired in 2007 by the world’s largest seller of rooms
online, the Priceline Group. Agoda.com is now a truly global enterprise
offering accommodations around the world, with offices in 50 locations
in 31 countries and more than 3,000 employees of 65 nationalities
(www.agoda.com).
Stakeholders Served
Stakeholders served are typically leaders, often at the top of the organization.
The most successful workforce analytics teams have the support of these
leaders. For some organizations, such as financial services and technology
firms, the core business is often built on a foundation of analytics. In some
cases, the chief executive officer (CEO) or the chief human resources officer
(CHRO) has a proclivity toward analytics. For these organizations, support
for workforce analytics comes easily. For others, securing support could be
more of a challenge. An important stakeholder management task is
determining your leaders’ views on workforce analytics and planning your
communications accordingly.
“It was less that we did an analysis and showcased it; it was more
proactively answering questions with data from our senior leaders. The CEO
and CHRO are very data driven.”
—Mariëlle Sonnenberg
Global Director, HR Strategy & Analytics, Wolters Kluwer
******ebook converter DEMO Watermarks*******
Understanding the views of your CHRO is particularly important, and the
best way to do so is through direct conversations. Al Adamsen, Founder and
Executive Director of Talent Strategy Institute, describes an experience early
in his career: “I was three levels down from the CHRO and four levels down
from the CEO. It was like playing a game of telephone tag—very frustrating
from an analyst point of view. You never know what’s really needed when
you’re just relying on messages being passed down.”
An organization’s board of directors can also play a decisive role in bringing
an analytics focus to HR. As discussed in the later section, “Board of
Directors,” human capital is growing as an area worthy of board attention.
Of course, you need buy-in and commitment from your organization’s HR
leadership and HR business partners; in many cases, they will be the liaisons
to the business leaders. With the HR stakeholder group, strive to create and
nurture an analytical mind-set. This does not require a detailed tutorial on
statistical techniques; instead, it involves building an appreciation for the
value that analytics can bring to the HR function. And make it personal:
Convey how analytics can improve stakeholders’ own contributions and
effectiveness.
HR Leaders
Because workforce data reflect information about people, and the work of HR
focuses on people, HR leaders should be integrally involved in workforce
analytics efforts. They might be on the receiving end of workforce analytics
output, they might sponsor a project, or they might be the ones expected to
act on the results. Andre Obereigner, Senior Manager of Global Workforce
Analytics at Groupon, says the role of HR in his work included addressing a
retention problem analytically: “Once I had a valuable working solution, we
showed it to the regional HR directors and they were quite excited. The
people using it now are the worldwide, regional, and country HR managers.”
HR leaders need to be part of the analytics process, beginning with issue
identification and hypothesis formation, all the way through to data
collection, analysis, recommendations, and action planning. Buy-in and
commitment to act are best achieved by working together throughout the
project lifecycle.
“Most HR people didn’t join HR because of an interest in analytics. To help
them to think analytically, we get them involved in a project that shows the
******ebook converter DEMO Watermarks*******
value of analytics—first-hand involvement helping them solve a problem
that’s relevant for what they do.”
—Thomas Rasmussen
Vice President, HR Data & Analytics, Shell
“The reason analytics has stalled in HR in the past is because there weren’t
any business stakeholders. If there aren’t business stakeholders, HR analytics
is interesting but not essential.”
—Josh Bersin
Principal and Founder, Bersin by Deloitte
Board of Directors
Some organizations are experiencing an increased interest in workforce
analytics from their board of directors. Given the influence boards tend to
have on organizational leadership, understanding this perspective is critical.
Laurie Bassi of McBassi & Company says, “Boards are asking more piercing
questions about HR issues that go deeper into the organization, beyond the
executive layer that has previously been the focus. Generally, boards are
paying much greater attention to their fiduciary obligations, and they are
realizing that the people in their businesses represent an asset at risk that
needs to be actively managed. They see human capital as a topic worthy of
board attention.”
Proactively investigate the board’s views and gain an understanding of the
board’s responsibilities regarding people-related topics (for example,
succession planning, executive compensation, and employee engagement).
The CHRO (or other senior HR leaders) can be a good source of the board’s
******ebook converter DEMO Watermarks*******
views of workforce-related topics. Your organization’s annual report, as well
as the annual reports of other relevant companies, can also provide useful
information. This insight can help shape the workforce analytics agenda and
will certainly strengthen your understanding of your leadership’s perspective
and expectations. Also talk to your CHRO about his or her specific
accountability to the board and, if you have the opportunity, work with the
board directly.
Data Owners
Identifying the people who can grant you access to the different types of data
you need for your analyses can be challenging. You might be able to work
out with senior leaders the types of data available, but actually finding the
person who can get you the data and enable you to work with it can be
difficult. Nonetheless, this is a necessary step. When you identify the right
people, you will need their cooperation to get what you need, when you need
it.
“People who work with business data are key. They know how to interpret
data, find possible bias in the data, and explain the results together with HR.
It’s very tempting to think you know the data when you get the first data
dump, but you need to know it well to understand it. The subject matter
experts in the business can help you with this.”
—Patrick Coolen
Manager, HR Metrics and Analytics, ABN AMRO
Keep in mind that the people managing these data sources are probably
completely occupied with their full-time job responsibilities. You want to
Technology Owners
The tools available to analytics teams are usually chosen, implemented, and
managed by a Human Resources Information Technology (HRIT) function or
an enterprise-wide technology function. When this is the case, you want to
form positive working relationships with the IT decision makers. This could
give you an opportunity to influence future technology purchases, as well as
alert you to any system-related scheduling and timing constraints. For
example, if a system that you need for extracting data is scheduled for a
maintenance window and will be unavailable for a period of time, work that
into your timeline so you don’t find yourself without a dataset at exactly the
******ebook converter DEMO Watermarks*******
time you need it.
Legal
HR datasets generally contain sensitive information, including personal
details about employees’ work history, performance evaluations,
compensation, and national identification codes such as Social Security or
National Insurance numbers. Some of these data elements are considered
sensitive personal information (SPI) in the United States and have similar
designations in other countries. This is information that, if compromised or
disclosed, could result in substantial harm such as identity theft.
“On every single project, we encounter legal issues—for example, getting the
right access to data. Engage your legal team as early as possible.”
—Andrew Marritt
Founder, OrganizationView
Given the personal nature and potential risk associated with handling HR
data, extra care and precautions are strongly advised. Work with your legal
department for guidance, advice, and counsel. Ensure that you understand
and are adhering to your organization’s policies and practices, and stay
mindful of country-specific regulations regarding employee data.
Also be sure to recognize that data sensitivities do not represent an
insurmountable obstacle. Understanding and adhering to guidelines will not
hold you back, but will instead help you to operate properly and securely. If
your organization has a chief privacy officer, this is an important person to
rely on for successfully navigating data security and privacy requirements.
Finance
Aligning with the finance department on the value of workforce analytics is
an important success factor. If you are able to anchor your data definitions to
those of finance, you will increase your credibility and be able to provide
insights that finance and business leaders find valuable. For example, after
you’ve identified the metrics that the business and HR focus on, you can
work with finance leaders to ensure that these metrics and their definitions
are sound. Mariëlle Sonnenberg, Global Director of HR Strategy & Analytics
at Wolters Kluwer, did this and also obtained the actual data from the finance
group. This approach allowed Mariëlle to focus on a targeted set of key HR
metrics that were most valuable to the business.
“If finance is not on your side, the workforce analytics function is not going
anywhere.”
—Placid Jover
Vice President of HR, Organisation & Analytics, Unilever
Stakeholders Impacted
By their very nature, workforce analytics projects will have an impact on
people. The individuals most likely to be affected are the very people whose
data are being analyzed (the workforce), the people who will be asked to
administer those changes (their managers), and the people who will model
and support the changes (their executives). Always keep these stakeholders in
mind when analyzing data and formulating recommendations based on the
results.
The Workforce
Workers are arguably the most directly affected by workforce analytics
actions. The workforce analytics team should never lose sight that each and
every worker is a unique individual, not merely a collection of data points.
“I do this work not only because it can drive organizational performance, but
also because it can improve people’s lives.”
—Al Adamsen
Founder and Executive Director, Talent Strategy Institute
The analytics might point to a very clear finding and recommended action,
but implementing that action might not be so straightforward. For example,
singling out a group of critically important employees and treating them to
special benefits might stem a specific attrition problem, but can you afford to
potentially alienate the 90 percent of the workforce that doesn’t receive the
benefit and is needed to keep the business operating? Implementing actions
for some while managing the effect on others is certainly possible, but ample
time and careful planning are needed to get it right.
Openness with employees about workforce analytics is recommended.
Organizations should disclose that they are managing HR with a systematic,
fact-based, equitable approach. Communicating the benefits should help allay
any concerns employees might have, says Mihaly Nagy, CEO of The HR
Congress and Managing Director of Stamford Global: “Probably there are
questions among employees about what HR analytics can do for them, what
benefits it can provide. With the right communication and education, this
resistance can be managed.”
Executives
Given their positional power and authority, executives play a particularly
important role in workforce analytics projects. In addition to being recipients
of workforce analytics outcomes, they are often relied upon to support the
recommendations in words and actions, to carry the message forward and
lead the way. Yet some executives might perceive an analytics project as
threatening to their authority. Laurie Bassi of McBassi & Company observes,
“Senior vice presidents sometimes have their own hypotheses and aren’t
always interested to have them debunked. After all, they’re doing pretty well
without analytics.”
Executives who have built successful careers with their current style of
managing and way of thinking might not be inclined to approach problems
from the same perspective as the analytics team, and understandably so. This
is an important possibility to be aware of because some executives might then
withhold needed support or block actions. If your team experiences this
situation, seek assistance from your project sponsors.
Figure 8.2 summarizes the various stakeholder types. The figure shows
recommended topics of conversation for engaging each group and the topics
relevant to specific stakeholder types.
Getting to Action
You can also influence the likelihood that actions recommended from your
analytics projects will be implemented. Reflecting on experiences early in his
career with previous employers, Ian O’Keefe, Managing Director and Head
of Global Workforce Analytics at JPMorgan Chase & Co. took certain steps
to encourage action. Ian computed statistical models to predict the outcomes
if actions were not taken and showed those scenarios to stakeholders: “We
showed trends, told them what would happen next week or next month if we
******ebook converter DEMO Watermarks*******
did nothing and what would happen over those time periods if we made some
changes. For example, we showed the overall cost of not managing poor
performance, in dollar terms. We showed them that this was pain that the
business didn’t need to live with.”
Stakeholder Responsibilities
Healthy relationships are two-way, and stakeholder relationships are no
exception. Stakeholders have responsibilities, too. Ideally, they will share
their knowledge and expertise, challenge the analytics team’s thinking,
suggest hypotheses to test, and be prepared to be surprised. Most important,
stakeholders need to take ownership of conclusions and actions, and they
need to acknowledge the full set of findings, not just those that confirm their
beliefs.
“We wanted them to commit to take action on the insight—we are strict about
that at the beginning. We also say you cannot choose the insights you work
on, you have to do all of them. Otherwise, they take only what’s convenient or
what they like.”
—Patrick Coolen
Manager, HR Metrics and Analytics, ABN AMRO
Summary
Working effectively with stakeholders is essential for workforce analytics
success. Relationships are needed with a wide variety of people throughout
your organization. Approach the task of stakeholder management
thoughtfully and systematically, with the following actions:
• Identify the various stakeholder groups, taking into account those served
by the analytics team, those you are dependent on, and those whose
working lives might be affected.
• Determine your stakeholders’ perspectives on analytics and adjust your
communications accordingly.
• Be cognizant of data privacy and security requirements, and seek legal
expertise to successfully navigate the requirements.
• Create, execute, and maintain a communications plan to structure your
interactions with stakeholders, including key messages and discussion
topics for each.
• Help stakeholders understand and fulfill their roles and responsibilities.
• Bring value to your stakeholders; make them more successful,
acknowledge their contributions, and listen to and address their concerns.
• Plan for the end from the beginning: Think through and discuss potential
analytics results and recommended actions, as well as how you will
evaluate actions that you implement.
Complexity-Impact Matrix
To help decide on your first project, a good approach is to plot the potential
project opportunities on a two-by-two matrix according to the level of
expected impact and the amount of complexity involved. Delivering a project
of moderate-to-high impact makes the most sense; a project that does not
have at least a moderate impact will likely go unnoticed. A project with low-
to-medium complexity is also a good candidate; high-complexity projects
take more time, and senior stakeholders might end up asking why your
project is taking so long.
Figure 9.1 shows the Complexity-Impact Matrix3 that results from this
exercise. Notice that it includes four types of projects: Quick Win, Big Bet,
Trivial Endeavor, and Pet Project.
Readers of this book are likely approaching workforce analytics tasks from
different perspectives. Some have a new role in a new function, others are
new to the role but are joining an existing function, and still other readers are
in the same role but with an expanded focus. Both experienced and
inexperienced practitioners can encounter each scenario. For these reasons, it
is important to consider these concepts relative to your level of personal
experience and your function’s history. For example, one team’s Big Bet
might be another team’s Quick Win. Similarly, a project that establishes an
analytics functions’ reputation in one organization might be a standard
project for a more experienced function.
Quick Win
A Quick Win is a project of low-to-medium complexity with moderate-to-
high impact. In short, this project is one that you feel confident you will be
able to deliver in a reasonable period of time and with tangible results.
Projects that involve uncertainty about the team’s capability and projects that
rely on dependencies over which you have little or no control are not Quick
Wins; they generally involve too much complexity to ensure a successful
result so early in the function’s existence. Projects that involve only a small
or moderate impact (for example, to the efficiency of an HR metric) also do
not constitute Quick Wins because the results are unlikely to get noticed
beyond HR.
Big Bet
A Big Bet project is a high-complexity project that is expected to deliver high
impact. Numerous factors can make projects complex. For example, you
Trivial Endeavor
A Trivial Endeavor project has low-to-medium complexity and low-to-
medium impact. If possible, avoid projects with lower expected impact in the
early phases of an analytics function’s development. The critical objective of
your first project is to deliver results that make a material difference to the
business and build confidence in your ability to undertake more projects.
Trivial Endeavors are unlikely to fulfill this requirement. Even though they
are not as desirable as other project types, it is useful to know what kinds of
projects fall into this category. For example, it can help to reframe your
project so that it resembles a more desirable project type, such as a Quick
Win.
The most common types of projects in the Trivial Endeavor quadrant are
those that lead to decisions that management can make just as effectively
without the use of advanced analytics. For instance, perhaps the company can
benefit from cost savings or additional features by switching providers of an
engagement survey or by shopping for a new supplier at the end of the term.
These problems are generally manageable using standard administrative
approaches.
Pet Project
A Pet Project is overly complex for the impact it delivers. These projects are
more aligned with personal interests than what the business requires. Avoid
Pet Projects: These projects (and Trivial Endeavors as well) won’t likely get
you the positive recognition you require to build the support you need for
your workforce analytics function. In fact, they could end up attracting
negative attention. To make matters worse, projects in this quadrant are
complex to execute because of factors such as the political environment, data
When it comes to impact, three broad factors play into rating projects as low
or high:
• Return on investment (ROI)
• Timing
• Opportunity cost
Assessing Complexity
Project complexity refers to the scope of the challenge you face in delivering
the project. Complexity is a relative concept. An advanced workforce
analytics function might consider a project low complexity, yet a nascent
******ebook converter DEMO Watermarks*******
workforce analytics function that lacks access to more advanced technologies
and resources might consider it high complexity. Therefore, your team must
always evaluate the project’s complexity in terms of its own capabilities and
the context in which you are operating. This section covers the five factors to
consider when assessing the complexity of a project for your team.
Assessing Impact
Impact is the level of benefit the business receives from undertaking the
analytics and implementing the follow-up recommendations. When
considering the expected impact of an analytics project, practitioners should
have three issues in mind: return on investment, the timing of the project
returns, and the opportunity costs of not undertaking other projects.
“Learn the logic of the business. What makes the business more successful?
How do people contribute to achieving this success? This is how you identify
projects that will make an impact.”
—Marcus Champ
Senior Manager, HR Analytics, Standard Chartered Bank
******ebook converter DEMO Watermarks*******
• Return on investment. The central aim of workforce analytics is to
realize business efficiencies and take advantage of opportunities.
Therefore, it is difficult to discuss the idea of impact in workforce
analytics without some discussion of cost. At its simplest, the issue comes
down to whether the expected cost of the project is less than the expected
return to the business from completing the project. This concept might
seem straightforward, but the cost decision is not quite so simple.
Managers must consider the cost of the project relative to the returns it
will deliver in relation to the next two factors (when precisely the benefits
will be realized and the opportunity cost of not investing elsewhere).
• Timing. The project’s timing issues can often be addressed by asking
yourself whether the project is focused on reducing costs or improving
productivity. For the most part, cost reductions are quicker to realize than
productivity gains. Therefore, focusing initial Quick Wins on cost
reduction might make sense. The impact of a project, like its complexity,
is relative to the business and the situation. In general, select a project that
will have a short-term impact; otherwise, it cannot really be considered a
Quick Win.
Although few people ever intentionally undertake an analytics project that
will not have an impact, this situation does happen. For this reason,
business executives must have clear insight into the project to make sure
that the improvements the workforce analytics team is predicting are
clearly tied to business expectations.
• Opportunity cost. When considering the impact of a project, it is
important to realize that the evaluation must occur in the context of other
possible workforce analytics projects—and also in the context of other
possible business projects. Even highly appealing projects that are
seemingly low complexity and high impact might be ranked behind other
projects when all options are considered. For this reason, it is important to
simultaneously consider several projects for impact, in case this process
reveals that another project is even more attractive than one you are ready
to initiate.
Summary
Selecting your first workforce analytics project can be a difficult challenge,
but taking a systematic approach to considering both complexity and impact
******ebook converter DEMO Watermarks*******
ensures that you make the most appropriate choices. In particular, remember
the following guidelines:
• Spend enough time planning your project to address all the hurdles you
expect to encounter, but be prepared for the unexpected hurdles that will
invariably arise.
• Identify potential projects that relate to the organization’s key
performance indicators.
• Classify your projects according to their complexity and their expected
impact, and go for a Quick Win that is low-to-moderate complexity and
moderate-to-high impact.
• When rating the complexity of the project, consider the following factors:
politics, skills, data, technology, and ease of implementation.
• When rating the expected impact of your first project, remember that the
project should deliver a sufficient return. The benefits should also be
realized in the short to medium term and should offer a greater net return
than investing in any other workforce analytics project.
Data are the very foundation of analytics. Without data, there are no
analytics. As we emphasized in Chapter 7, “Set Your Direction,” the
starting point for analytics should always reflect your vision and
include a clear articulation of what you are striving to accomplish.
However, achieving that vision and mission is not possible without
data.
This chapter covers the following:
• A practical approach to data
• Data challenges and solutions
• Types and sources of data
• Data governance
“The biggest challenge? Data. Not many organizations have a global data
warehouse. Data aggregation, data cleansing, having a single trusted source
—these are the things we spend most of our time on with clients, not analysis.
The latter turns out to be straightforward once the data are in shape.”
—Michael Bazigos
Managing Director and Global Head of Organizational
Analytics & Change Tracking, Accenture Strategy
Missing Data
Suppose you want to determine whether experiences at work differ for men
and women. You might choose to conduct a survey to answer this question.
Several participants in the survey might choose not to answer the question
asking them to indicate their gender, resulting in output like Figure 10.1.
With missing data such as this, you must determine how to proceed with the
analysis.
Outdated Data
You might have access to the data you need for your analysis, but some of
the values might not be up-to-date. As an example, suppose you want to
determine whether compensation is related to productivity (to find out
whether more highly paid people produce more). To do this analysis, you get
a data extract from the organization’s core Human Resources Information
System (HRIS). You learn that the dataset does not reflect recent off-cycle
salary increases because the compensation system (which records salaries)
has not yet synchronized with the core system. If you are able to source this
information only from the HRIS, the current salary will be incorrect for the
people who were part of this off-cycle salary program. You then need to
determine the implications for your analysis.
Judgment is needed to determine how much of an issue this is for your
analysis and conclusions. If the number of outdated values is large enough to
appreciably influence the outcome of the analysis, you will want to make
every effort to obtain updated information. A sensitivity analysis can be
helpful in determining whether updated values will significantly change the
findings. If updates are required, work with data owners to identify the best
option (some of which are described here).
If data refresh cycles are frequent (or imminent), your best course of action
might be simply to wait for the next refresh. For example, if the
compensation system feeds the core HRIS monthly and you need the very
latest compensation data, find the specific update schedules and determine
whether it makes sense to wait. If waiting for a refresh does not fit with your
timeline, you might be able to obtain access to the data from a different, more
updated source for specific variables (for example, the source compensation
system itself); then you can merge this data extraction with your master
dataset.
Another option is to manually update values, if necessary. If you opt for this
No Data Available
Sometimes no systems or databases have the data you need for an analysis.
Imagine that employees in a specific part of the business are quitting their
jobs at a high rate, and you need to determine the cause. You have several
factors to consider, but perhaps the sponsor of the project is particularly
interested in looking at promotion history (that is, when and how often people
have been promoted to the next job level). However, you learn that no system
has recorded promotion information. The data you need for the analysis
simply do not exist.
Does a lack of data mean that you cannot consider this variable in your
analysis? As with other data challenges, one solution is to initiate a new data
collection effort. However, you might be able find a better option with a bit
of creativity and ingenuity: You might be able to approximate the data you
need by using a combination of variables that do exist. For example, if you
need data on people’s promotion history, you could look for instances in
which people had a title change and a corresponding salary change. If
someone got a new job title and a raise at exactly the same time, this
combination of events is a strong indicator of a promotion and can be used to
create what you need without any incremental data collection. Another
creative approach is to consider external publicly available data as a proxy.
For example, this could be represented by job title changes posted on
LinkedIn.
Why does this matter? If you use statistics that require normality, but this
assumption is not met, the statistical tests could yield misleading results. The
whole point of statistics is to build a fact base on which to inform decisions;
if the fact base is inaccurate because analysis tools were misused, it runs
counter to that goal. As described earlier, tests are available to determine
whether your data meet the normality assumption. If you suspect they do not
(see Figure 10.3 for another example of a non-normal distribution), you can
either apply corrections to approximate normality in the data or use
alternative statistical techniques for analysis. Experts such as data scientists,
analysts, and industrial-organizational psychologists (within your team, from
other functions in the organization, or outside partners) can offer guidance on
how best to proceed.
Data Outliers
Outliers are values that are abnormally higher or lower than most other values
in a sample of data. Identifying outliers in your data is important because a
few extreme values can alter the results considerably. Sometimes outliers are
legitimate values; other times, they are the result of a data error. Either way,
they can lead to misleading conclusions. Tests are available to measure
outliers. At a minimum, always examine the distributions of your data as a
check before you run statistical tests (see the examples in Figure 10.4 to get a
sense of what to look for when plotting data graphically). If you identify
outliers, you need to make an informed decision on whether to include or
exclude them in the analysis. Including extreme values might mask an
important relationship or insight. Excluding them might hide a meaningful
variation. Consult the data owners to help with this decision.
Data Governance
After the first successful analytics project has been implemented and
recognized, it’s important to take the time to invest in data governance. Data
governance refers to comprehensive strategies, policies, standards, and rules
for managing data in your organization. This includes decisions and
******ebook converter DEMO Watermarks*******
agreements on all things data—what data elements to measure and store, how
to define each element, who is responsible for integrity and maintenance,
who can access the data, and more.
“Data governance has become a hot topic in HR. Ten years ago, it wasn’t a
consideration. Data quality is recognized as important today, and it’s
essential in getting to a level of analytics maturity with defined and
repeatable systems.”
—Jeremy Shapiro
Head of Talent Analytics, Morgan Stanley
Summary
Data are essential building blocks for workforce analytics, and relevant, high-
quality data are needed for quality results. The following guidance helps you
strike the right balance between ensuring data quality and progressing the
workforce analytics agenda:
• Build relationships with data owners to facilitate data access and learn the
details of the datasets (such as allowable data values, methods for
interpreting the data, and ways to spot errors); utilize data profiling
technology to assist with data checking.
• Check data for missing values, determine reasons for the missing data,
and take appropriate corrective action (drawing on expertise as needed).
• Verify that you have the most current and complete version of the data
needed for the analysis.
• Determine whether you can create the data you need from the data you
have, or find a proxy for the data you need.
• Consider a full spectrum of data sources and choose those that best
answer your questions.
******ebook converter DEMO Watermarks*******
• Take the time to establish data governance processes, including data
steward roles, to ensure ongoing data quality.
• Participate in online forums and user groups to stay current with the latest
views on data challenges and solutions.
• Hire or partner with a data scientist to assist with data decisions; an intern
can be a cost-effective approach.
• Recognize that data collection and analysis will be iterative and that you
will refine and improve as you go.
For example, if your analytics mission is in its early stage and you have both
a limited team and limited technology, you might be working mostly with
spreadsheets to report workforce data. For relatively simple analyses, this
could be fine, but spreadsheets struggle to cope as the volume of data and the
demand for analytics across the business increase. Eventually, you should
switch from basic spreadsheet software and consider what other technology
you need to profile and interpret the data. Consider what tools and services
enable better understanding and aggregation of the data than standard desktop
spreadsheet software such as Microsoft Excel.
Other organizations that have been working with analytics for a long time
have a very different starting point than those that rely mostly on Microsoft
Excel. They might have already built a data warehouse and could be asking
questions such as “How can we get more value from our data?”
Considerations for these organizations depend on the business problems
inspiring their analytics efforts. For example, they might be interested in
sharing their data with the rest of the HR function through some kind of self-
service provision. These organizations also need to understand their current
challenges and the technology that will help them meet those challenges.
Working with a partner on your technology strategy is worth considering; we
discuss this later in this chapter in the section, “Technology Vendor
Relationships.”
******ebook converter DEMO Watermarks*******
A MIND-SET FOR TECHNOLOGY
At ANZ Bank,1 Kanella Salapatas has had a great deal of success
winning over people to the possibilities of data and analytics
technology. Kanella is the HR Data Manager and Reporting Service
Owner, and her trick has been to get people involved with the
technology throughout the journey.
According to Kanella, the value in involving HR people with
technology comes from the confidence they gain when they know
what’s going on with the data and the analyses. “If a business leader
asks why a certain number is 18 and not 20, for example, the HR
Business Partner (HRPB) must have an answer and must be confident
about it.” Kanella has found that some HRBPs are reluctant to use
technology to really understand those answers.
To overcome such technological reluctance, Kanella has taken a two-
fold approach:
• First, she helps train people in basic skills of using technology such
as Microsoft Excel or business intelligence toolsets. Learning how to
query and explore information for themselves builds confidence in
the technology solution and people’s knowledge of where they can
get an answer when put on the spot by a business leader.
• Second, Kanella gets people involved in creating the reports they
receive. “Getting HRBPs involved in the construction of their reports
means that they will know the formula for the metrics and the
individual data elements used. In turn, this will allow them to more
confidently have conversations with their business leaders, and this
gives them additional credibility.”
Kanella’s advice is simple: Involve your HR leaders and HRBPs in
your technology and take away the mystery.
HR Data Warehouse
Most HRIS platforms from major providers are built around a vendor’s data
warehouse technology. However, in many cases, organizations using these
HRIS platforms find that they do not provide the flexibility to handle the
variety of analytics they want to undertake. Organizations with multiple
HRISs are also interested in linking their data across systems.
For these reasons, many organizations take regular data exports from
proprietary vendor systems and store the data in a specially constructed HR
data warehouse. These periodic data exports form the system of record used
to populate all other downstream systems that need accurate data for either
advanced analytics or HR reporting purposes. The data from the operational
“On a monthly basis, we export selected parts of our HRIS data and store it
in our wider data warehouse. We can overlay the definitions that we require
for particular metrics and the specific hierarchies we need for data mapping
rather than having to rely on how the HRIS defines things. We combine these
data exports with such things as customer research data and can do all sorts
of useful analytics on it.”
—Sally Dillon
Head of Business Intelligence, UK Life, Aviva
Reporting Technology
Reporting technology in workforce analytics refers to business intelligence
(BI) software applications that sit over the top of the HRIS, the HR data
warehouses, or both. The HRIS provider can supply this functionality in an
******ebook converter DEMO Watermarks*******
integrated single system, but many third-party vendors also offer reporting
technology to complement the large HRIS provider systems. Reporting
applications allow users to undertake tasks such as querying the HRIS for
information about each of the HR subfunctions. Types of reporting that can
be carried out include generating payroll reports for tax reporting
requirements, examining the average length of time taken to fill open job
roles, and checking the level of engagement with online learning
management systems.
BI technology can also be used to regularly examine how well the HR
function is performing relative to a set of metrics or key performance
indicators (KPIs). These tasks often involve using reporting technology to
populate dashboards or generate scorecards with information on topics such
as HR compliance violations, turnover rates in different parts of the business,
and numerous other HR metrics.
DRILLABILITY IS KEY
Sally Dillon is Head of Business Intelligence at UK Life at Aviva,2 a
large insurance company based in the United Kingdom. She has been
the data lead on large systems implementation projects, such as
Workday, to help HR get the most from their people systems. One of
her key approaches is to make sure the technology allows for
drillability, or the ability for users to thoroughly explore their data at
increasingly fine levels of detail.
For most HR professionals who lack highly technical skills,
drillability is possible only with a strong user interface to the
technology. “We encourage people to understand their data, and that
means having a very usable technology,” explains Sally. “The
interface must be user friendly. The technology should also have
strong mobile capabilities, since many HR professionals rely on their
mobile device to access information. Finally, it must have connectors
into the enterprise people systems and the data warehouse.”
Keeping her user’s hat on at all times has helped Sally ensure that
Aviva’s technology choices are delivered for even the least
technologically minded: “It’s not only about the user interface,” Sally
reveals,“You also need to put the data in the language of the business,
Cognitive Technology
Cognitive computing is a new area of technology based on artificial
intelligence intended to help advise HR. This technology learns, reasons, and
advises the user based on models that the system builds and inferences that it
makes. The user can interact with the system through natural language and
apply the models that the system builds. Cognitive computing technology is
playing an increasing role in analytically enabling HR practitioners.
Visualization Technology
Visualization technology refers to software used to generate graphs and other
******ebook converter DEMO Watermarks*******
visual representations of data. Visualization in analytics serves several
purposes.
First, visualization can help analysts gain a clearer picture of the data they are
considering. By their very nature, summary statistics such as means and
standard deviations convey limited information about the data they
summarize. For example, the mean does not tell you how spread out the data
are for a particular variable. Even summary statistics developed to describe
variation, such as the variance, will not readily spot outliers. Graphs can
easily highlight the average and the variability in data. See Chapter 10,
“Know Your Data,” for a discussion of why issues such as data variability
and outliers matter in analytics.
Second, visualizations help those who have a hard time interpreting data
understand the results of analyses. People who are less familiar with
interpreting statistical information often find graphical representations more
intuitively understandable. For this reason, visualization is an important part
of the insight generation and communication processes in analytics.
Finally, visualization can reveal complex results. Important developments in
statistical modeling and machine learning allow complex statistical
relationships to be developed. These cannot easily be represented graphically
using basic presentation software. Visualization software can bring these
complex relationships to life.
Share
Before you decide whether to lease or own technology for workforce
analytics, it is important to assess whether having exclusive access to the
technology is necessary for your workforce analytics goals. If you need
access to the technology on a more limited basis, you might be able to
leverage existing technology elsewhere within the organization. An example
might be access to high-powered computers for complex analyses or access
to technologies that distribute processing tasks across multiple idle computers
throughout the organization. If you do not need immediate access to such
technology, you might be able to share technology—that is, access existing
technology that other functions, such as marketing, use. However, if you
require access to technology on demand, you will likely want to lease or own
your own systems.
Subscribe
Many technology vendors now offer the option to subscribe to their software
through Software as a Service (SaaS) licenses. Other service subscriptions
include Infrastructure as a Service (IaaS) and Platform as a Service (PaaS).
These cloud technologies have important implications for workforce
analytics. First, cloud technology removes the need for significant upfront
capital expenditure (CAPEX) by turning it into a monthly subscription
(which is appealing for cash flow reasons as well as reduction in CAPEX). In
addition, with cloud technology, HR typically does not need to involve the IT
department in either the purchase decision or ongoing support.
Subscription approaches have the advantage of distributing the cost of the
software over time. They also disperse some of the responsibility for
maintaining the capability and performance of the hardware and software to
the technology vendor. Technology vendors that lease technology often give
service agreements, which means that while the lease is current, your
organization has right of access to any updates to the technology that the
Own
If the function requires exclusive access and you are not concerned with the
rate of renewal and refreshing of service capabilities, buying the technology
outright might be a sound option. In general, buying technology makes sense
if important new functionality is unlikely to emerge and if the technology
will be needed longer than it takes the asset to depreciate financially.
Technology that typically falls into this category includes hardware for
analytics, such as powerful desktop computers and servers. Although new
and faster technology will certainly come out, the desktop machines and
servers you purchase for your function will remain capable of performing all
the tasks you require of them for the foreseeable future (albeit not as fast as
newer, more powerful machines). Of course, ownership entails upfront
capital expenditure, although the value depreciates over the life of the asset.
Vendor Selection
If your workforce analytics function determines that leasing or buying
equipment is the best approach, you need to select the best vendor. Most
organizations have technology policies dictating whether an open competitive
tender is required. The key point is that workforce analytics professionals
need to influence the direction of tender processes so that decisions support
the goals of workforce analytics.
Involvement in the tender process will include writing the tender brief,
answering requests for information, and selecting the vendor. At a minimum,
******ebook converter DEMO Watermarks*******
workforce analytics professionals need a good working relationship with
procurement teams because their expertise is needed to guide the tendering
process for workforce analytics.
Summary
Decisions about technology are among the most important ones you will
make in workforce analytics. To make the right decisions, you need good
advice from Information Technology (IT) and HR Information Technology
personnel and from your data scientists. To prepare for these discussions,
take the following steps:
• Decide on the appropriate mix of technology by understanding what you
have now and what you need to perform the required types of analyses in
support of your workforce analytics vision and mission.
• Get familiar at a high level with the basic components of workforce
analytics technology, including the benefits of cloud technology.
• Consider whether and when cloud-based services will work in your
organization.
• Understand the breadth of technology required for a workforce analytics
technology system: the HRIS, the HR data warehouse, reporting and
business intelligence technology, analysis and data integration software
for advanced analytics, cognitive solutions, and visualization software.
• Decide whether to share, subscribe to, or own your technology, based on
ease of access, financial cost, and your degree of concern over how the
technology ages.
1 ANZ Bank traces its origins to the Bank of Australasia in 1835. Today it
is one of the five largest companies listed in Australia and is the biggest
bank in New Zealand. As of April 3, 2017, it had over nine million
customers and more than 50,000 employees. ANZ Bank is headquartered
in Melbourne, Australia (www.anz.com).
2 Headquartered in London, U.K., Aviva is an insurance company with 33
million customers and approximately 28,000 employees in 16 markets in
the United Kingdom and throughout Europe, Asia, and Canada. It has a
rich heritage dating back to 1696; one of its most famous customers was
******ebook converter DEMO Watermarks*******
Winston Churchill, who took out a policy in 1896. In its 2015 annual
report, Aviva reported revenues in excess of £23 billion
(www.aviva.com).
“There are so many different skill sets needed, from data infrastructure
through to reading literature, forming hypotheses, collecting data, doing the
analysis, and implementing recommendations.”
—Mark Huselid
Distinguished Professor of Workforce Analytics, Northeastern University
Skill 2. Consulting
Consulting skills include the ability to provide specialist expertise to
organizations to improve some aspect of their business. Core components of
consulting skills in the context of workforce analytics include the ability to
define business problems, generate hypotheses about the causes of business
problems, manage projects, develop solutions, manage stakeholders, and
manage change.
• Problem definition. Problem definition refers to accurately
understanding the nature of a problem. Doing this well requires skills with
language, such as listening, questioning, and paraphrasing, as well as the
ability to exercise good judgment. These skills are required in the first
step of the analytics model that Chapter 4, “Purposeful Analytics,”
outlines. Problem definition is an essential skill for workforce analytics
success because it allows you to not only clarify the business problems,
but also identify the most important ones. A good way to acquire skills in
defining problems is to consider case study materials about business
problems (for example, in business journals such as Harvard Business
Review). You can aim to apply the same approaches and thinking to your
own situations.
• Hypothesis building. Hypothesis building involves identifying plausible
causes of the problems you have defined and, more important, generating
potential courses of action that might solve the problem. Being able to
“When you are clarifying and building hypotheses, I have come to realize
that one of the most useful skills is to have what you could call a beginner’s
mind-set so that you can tackle every problem in an open-minded way.”
—Salvador Malo
Head of Global Workforce Analytics, Ericsson
Skill 6. Communication
As your team begins to build momentum, your team members will need to
develop or gain access to communication skills. These skills are critical to
effectively create and tailor messages, and they involve the following key
areas:
• Storytelling. Storytelling is a method of explaining a series of events
through narrative. Telling an effective story to capture the essence of the
problem, analysis, insights, recommendations, and change helps you gain
buy-in for your ideas and projects. As analytics have become more
important in the business environment, so have the skills required to
explain data and insights using a compelling narrative. Chapter 17,
“Communicate with Storytelling and Visualization,” covers this topic.
• Visualization. Skills in data visualization include the use of technology,
graphics, and artistry to clearly show insights. These skills often require a
******ebook converter DEMO Watermarks*******
combination of artistic creativity and technical ability with visualization
methods and tools. These skills are relatively new in the world of
workforce analytics; Chapter 17 discusses this topic, and Cole
Nussbaumer Knaflic’s book Storytelling with Data (Wiley, 2015) is a
comprehensive resource.
• Presenting. Creating and communicating your story through a structured
presentation is an important and much-needed skill. The presentation
might be informal and one-to-one, or it might be formal with many people
in attendance. It might be in person, or it might be delivered remotely.
Being able to adjust your presentation to the medium and audience is a
critical skill if you are to communicate your ideas in a compelling way.
For example, a slide deck should not be distributed without an
accompanying explanation. Presentation skills can be acquired though
experience and often via internal corporate training programs. All
analytics professionals should receive formal training in presentation
skills and basic presentation software such as Microsoft PowerPoint.
• Written communication. Clearly articulating written messages is
essential in workforce analytics. This is because you are often dealing
with complex ideas, diverse audiences, and different communication
media (for example, email versus extended documents). Succinct written
expression of ideas contributes to a clarity of understanding between your
team and others. For highly important documents, you might choose to
use writing and media specialists to improve the clarity of your messages.
• Marketing. Marketing is about communications that persuade an
audience that an organization’s brand, products, or services have value.
As you seek to persuade others of the benefits of your analytics team’s
work, either internally or externally, you will want to market your team
well. This includes managing your team’s brand and its reputation.
Effective marketing often requires the services of a marketing
professional, whether within your organization or from an outside
specialist.
Summary
Building the analytics team or function is a critical task in the early phases of
workforce analytics because the capability of your team is an important
determinant of your success. This can seem like a challenging task, but taking
a structured approach and following the principles outlined in this chapter
sets you on the path to success. Be sure to cover these tasks:
• Ensure that you understand and have access to the Six Skills for Success,
whether they are on your team or are easily accessible elsewhere.
Configure the skills in your team and its size based on the expected
workload and nature of projects.
• When deciding whether to hire new staff or develop existing staff, be
aware of the pros and cons of each approach, such as whether skills
developed in another area transfer to a workforce analytics context.
• Remember the fundamentals of personnel selection. Cognitive ability and
certain personality traits, particularly conscientiousness, are important in
all roles and can be tested with standardized psychometric testing.
Advantages of In-house
The in-house approach for skills offers several advantages, such as control
over the nature and timing of projects. Having resources all contained within
a single team makes it relatively easy to reassign people to projects as
needed, adjust project timelines as business demands evolve, and facilitate
collaboration and knowledge sharing among team members. Another
advantage of the in-house approach is the opportunity to invest time (for
example, when in-between projects) to innovate and explore new statistical
methods and novel data sources.
Disadvantages of In-house
Potential disadvantages of the in-house approach include needing to make an
upfront investment in people and the tools they need to do their jobs. This
requires in-depth planning for decisions with long-term implications; after
committing to a team of people and purchasing a set of tools, backtracking is
difficult. Staffing an in-house team also involves a lengthier initial startup
time as you source candidates, bring new hires on board, train them and get
them up to speed, and put tools and methodologies in place. The leader of a
fully staffed in-house team can expect to spend a large proportion of time
managing the team, with less time for managing projects and stakeholder
relationships.
Sizable investments are often accompanied by sizable (if not outsized)
expectations, and that can be the case when building an in-house analytics
team. Expectations of what the team can deliver according to certain
timelines could be unrealistic, leading to disappointment. Some practitioners
operate with a low profile early on to avoid this issue. Their preference is to
surprise stakeholders with valuable insights and results, as a way to garner
further support.
Another factor to consider is the political environment in the organization,
specifically in the context of challenging long-held assumptions and being
able to effect change. In some organizations, an internal team might lack the
credibility that comes from (perceived) third-party objectivity. Also relevant
is whether HR is seen as having analytical prowess instead of being thought
“The risk modeling division inside the bank are very good partners. They
have made fraud calculation predictions to very advanced levels. We use
their data scientists when we need them. They are very good, they already
predict customer attrition, so we just ask them to apply their models to
different variables.”
—Kanella Salapatas
HR Data Manager and Reporting Service Owner, ANZ Bank
Advantages of In-sourcing
In-sourcing skills offer several advantages, such as the depth of knowledge
that internal experts have about the organization, in addition to their subject
matter expertise. Internal experts bring a level of understanding about goals,
challenges, operations, and systems that will be unmatched by an external
provider or a newly hired team member. And somewhat similar to an external
party, internal experts are less likely than an HR insider to have preconceived
notions about HR-related projects and can bring a fresh perspective (and
credibility) to the analysis.
In-sourcing is a cost-effective option that requires no substantial incremental
investment, as long as all parties involved are able to fulfill their collective
scope of responsibility. In-sourcing also extends the reach of workforce
analytics to a broader audience within the organization, helping to build a
coalition of supporters.
Disadvantages of In-sourcing
Among the disadvantages of relying on other people in the organization who
have their own commitments and responsibilities is the uncertainty associated
with team members’ availability, time, and focus to work on your projects.
Other projects might take priority for them at any given time, and resources
could be redirected, impacting your timelines and ability to fulfill your own
commitments (although this is not an issue when resources are formally
assigned to the HR team).
Another challenge is that, if the analysts are too far removed from the HR
function, a lack of domain expertise could potentially lead to inefficiencies
and rework. Without the HR “sixth sense” (see Chapter 12), the analysts
might identify relationships in the data and recommend actions, without
taking into account issues such as unfairness and discrimination.
Finally, borrowing expertise from elsewhere in the organization could give
the appearance that HR is not capable of delivering analytically based
insights. The in-sourcing option could potentially and inadvertently reinforce
this type of preconceived notion that exists in some organizations.
“Businesses don’t have two years to wait for you to build sophisticated
analytical capability. By tapping into external partners, you get the skill you
need immediately.”
—Mark Berry
Vice President and Chief Human Resources Officer, CGB Enterprises, Inc.
When chosen wisely, external providers will work with you as a true partner
in realizing your workforce analytics goals. As Andrew Marritt, founder of
OrganizationView, states, “The best consultants put themselves in the shoes
of their clients.”
Advantages of Outsourcing
In addition to providing a ready source of deep expertise, external partners
have the distinct advantage of providing a third-party perspective. Strong
partners are skilled at asking provocative, constructive questions that might
be difficult for an internal person in a politically complex environment to ask.
Laurie Bassi, CEO of McBassi & Company, expands: “Our job really is
shifting mind-sets by focusing on the power of asking good questions. As an
outsider, this can be easier to do because we are not lost in the daily grind.”
Partners can also offer experience in overcoming obstacles and points of
resistance. Peter O’Hanlon, Founder and Managing Director of Lever
Analytics, says, “Partners can say the hard things and make things happen
quickly because they’ve seen it before and they know how other
organizations work.” Partners often bring repeatable approaches that have
worked well in other organizations. As a result, you can benefit from their
learning and experiences and thus accelerate the time to realizing value from
analytics. Michael Bazigos, Managing Director and Global Head of
******ebook converter DEMO Watermarks*******
Organizational Analytics & Change Tracking at Accenture Strategy,
describes his approach at a previous organization: “We built a team with
domain expertise as well as quantitative expertise, and we developed
replicable solutions to take to clients to solve problems fast.”
Skilled external partners can customize their support to meet the evolving
needs of their clients. They might be able to assume all responsibilities for
statistical modeling, for example, or they might be able to coach their more
advanced clients to continually improve their analytical skills. For those in
between, partners can do as Andrew Marritt of OrganizationView has done
and coach them to be a consumer of analytics—for example, teaching clients
what to look for, what questions to ask, and how to interpret results.
In addition, not every organization has the need or desire to staff a complete
analytics function in HR. Mark Berry of CGB Enterprises, Inc., asks, “Why
build something internally when you can ‘buy’ deeper expertise on the
outside? As an organization, you don’t have to build internal capabilities for
every area of HR analytics. You can let expert service providers do it for
you.”
Disadvantages of Outsourcing
Although utilizing external providers allows a degree of flexibility in
spending, the cost can exceed that of having one or more full-time team
members if the team uses external providers extensively. Another potential
challenge is lack of continuity: Specific individuals at partner organizations
might not always be available for your projects if they are in high demand by
other clients.
External providers also need time to learn about your particular
organizational culture and environment, the specifics of your data, and the
details of your organization’s operations. Data security might also be a
consideration because you will need to release your sensitive people data for
analysis. In addition, you will likely have less flexibility to adjust your efforts
to accommodate changing priorities. Organizational demands can and do
shift, sometimes in the midst of a project. Keep in mind that you will still get
billed for time spent on paused or halted projects, with limited, if any, value
to show for it.
Table 13.1 summarizes the advantages and disadvantages of the three broad
options for building workforce analytics skills: in-house, in-source, and
outsource.
Budgets
******ebook converter DEMO Watermarks*******
Some organizations have the good fortune of generous budgets for building
workforce analytics capabilities, but most face budgetary constraints. The
amount of funding available defines the boundaries of what can initially be
built. If you are in the fortunate position of having sufficient funds available,
you might opt for fully staffing an internal team and supporting it with state-
of-the-art tools and technology. As noted in the following sections, however,
budget is not the only consideration. Teams with more limited budgets need
to estimate the cost of each best-fit option and choose the path that conforms
to their budget. As you execute your first few projects and demonstrate
success, you will likely have further opportunities to build a business case for
a larger investment.
Organization Size
The size of an organization partly dictates the approach for building
analytical capability. Small organizations are likely to benefit most from
outsourcing anything beyond the minimum capabilities. This is because, for a
small organization, benefits are unlikely to scale to the same extent as in
larger organizations. In addition, partnering allows small companies to
connect with highly skilled experts when they need them, without the burden
of large salary expenditures and lengthy ramp-up times. Mark Berry explains,
“I don’t have to maintain those sunk costs, and I can manage the cyclical
nature of the business. The outsourcing model is the right model for us.”
Perception of HR
Another consideration is the internal perception of the HR function. In some
instances, lines of business develop their own HR analytics capabilities
separate from the HR function. This can occur because of a perception that
HR is not meeting their business demand. For example, a business needing to
hire a large number of staff in a short period of time might have prompted
those business areas to develop their own analytical techniques for
******ebook converter DEMO Watermarks*******
forecasting workforce requirements and filling vacancies.
If you encounter such a situation, a collaborative approach is recommended.
Listen carefully to your colleagues’ needs, and demonstrate your team’s
analytics know-how and willingness to experiment and take risks. If this
situation arises an in-source approach, partnering with the line of business, is
recommended. In most instances, the best outcomes result when the lines of
business and HR work together to address business challenges.
Third-Party Objectivity
Some organizations use external parties because of their impartiality and
perceived credibility. In addition, some projects are sensitive in nature and
preclude an internal team from working on them. For example, leadership
analytics for selecting potential CEO successors would require restricted
visibility, as would analytics related to a merger, acquisition, or divestiture.
In these instances, partnering with an external provider will likely be
necessary.
Partnering allows flexibility in acquiring the skills needed to fulfill the
workforce analytics vision and mission. As your team evolves and your
positive contributions to the business become clear, you will have
opportunities to secure additional budget, source more skills, and, ultimately,
build the function that best serves your needs. Mihaly Nagy, CEO of the HR
Congress and Managing Director of Stamford Global, observes, “When
organizations don’t have the resources in-house, they turn to consultants to
help them on the journey. Once they see early wins, more funds become
available. Then dedicated functions get built.”
Summary
Establishing a workforce analytics function does not necessarily mean
staffing a fully skilled end-to-end in-house team. Although that is certainly
one option, collaborating with partners (either internal to the organization or
external providers) could be the best path to success. Consider the following
when choosing among your options:
• Understand the options for bringing together the needed workforce
analytics skills (in-house, in-source, outsource) and the advantages and
disadvantages of each.
As you can see at the top of Figure 14.1, an operating model keeps your day-
to-day operations consistent with your workforce analytics strategy, as
defined by your vision and mission. A governance framework helps you
balance the needs of various stakeholders, from specifying the proper
handling of data, to operating within the designated reporting structure, to
defining decision-making procedures. Within that framework, you need to
structure the implementation of your ongoing work through team structure,
clarity of roles and responsibilities, and a disciplined approach to project
******ebook converter DEMO Watermarks*******
management. And as with any business investment, accountability for results
is required, through articulating a business case and tracking metrics that
demonstrate your impact relative to those investments.
Strategy
To ensure ongoing relevance, it is important that the workforce analytics
team’s work aligns with and supports the organization’s overall strategy. As
business conditions change and the organization adapts, the workforce
analytics team’s focus should similarly evolve to meet new challenges and
requirements. Periodically verifying your team’s vision and mission will help
maintain the alignment needed.
Governance
Governance provides a framework for how a group operates within its larger
organization and environment. It is extremely useful for clarifying rules of
engagement when working with workforce data, as well as specifying
effective reporting relationships and guiding decision making.
Data policies and regulations exist for good reason: to keep everyone safe
from data misuse, fraud, and unethical and illegal activity. The increasing
availability of data and the increasing ease of analysis should not be used to
justify unconstrained and ill-conceived analyses. For example, organizations
might have the ability to collect information on marital status or the number
of children in an employee’s family, but using that information to make
decisions about employees would be inappropriate in some jurisdictions or a
******ebook converter DEMO Watermarks*******
violation of employment law. Furthermore, no organization should take
action on the basis of the marital or family status of its workers. Eric
Mackaluso, Senior Director of People Analytics, Global HR Strategy &
Planning, ADP, further points out, “It’s an ethical question; just because you
can do it doesn’t mean it’s the right thing to do.”
HR domain expertise is extremely useful in ensuring proper use of data.
Having this expertise on your team makes you better prepared to avoid the
pitfalls and dead ends that result from spending time on statistical
relationships that are inappropriate, potentially unethical, and possibly
irrelevant for consideration. As a cautionary tale, consider the case of a
multinational organization we encountered. In this example, an entire
analytics project was derailed because it pointed to recommendations that
were potentially discriminatory. The analysis was sound and the business
question was highly relevant to the organization, however, the team failed to
spot the warning signs until the storytelling phase with the executive
sponsors. At this point the project was halted. Upon reflection, the analytics
professional realized this outcome could have been avoided with proper
knowledge of HR practices surrounding the use of people-related data.
You must also be mindful that data collected for some purposes cannot
always be analyzed for other purposes. Even if you have what appears to be
the perfect dataset, you must respect legislation, regulations, and policies that
indicate it is inappropriate for further analysis. In cases such as this,
alternative data might need to be identified or new data collected. When
collecting new data, do so with an eye to the future analyses you may wish to
conduct and obtain suitable permissions accordingly.
Note that often it is perfectly acceptable to analyze existing data in aggregate
at the group level. For example, a retail company might want to understand
differences in performance across its different store outlets. Analyzing
variables such as employee survey scores at the outlet level should be
perfectly acceptable. The sensitivity tends to arise when data are used for
making decisions about individuals—specifically, decisions that were not
intended when the data were collected. For example, employee survey scores
designed to measure department-level engagement should not be used to
isolate particular employees with low scores for some sort of action (such as
managing them out of the business).
Finally, when collecting new data, you must be open and transparent about
******ebook converter DEMO Watermarks*******
analytical intentions. Teams can take steps to increase employees’ comfort
level with sharing their data for analytics purposes. We recommend a four-
factor framework developed by Nigel Guenole and Jonathan Ferrar in 2015,
identified by the acronym FORT:
• Listen to feedback on the analytics goals from people who will be
affected.
• Where possible, make sharing of data optional.
• Give some recognition to those who agree to share their data for
workforce analytics.
• Be transparent about everything that is done.
To summarize, the following actions are essential to ensure responsible and
ethical handling of HR data:
• Understand and adhere to all legal and regulatory requirements regarding
collection, usage, storage, and sharing of data.
• Understand and adhere to all company-specific data guidelines.
• Keep your knowledge of guidelines and regulations up-to-date through
relationships with information councils in your organization.
• Understand the boundaries of the appropriate use of people data.
• Strive for broad approvals for future data usage when undertaking new
data collection efforts.
• Establish an environment (based on gathering feedback, opting in,
recognizing contributions, and ensuring transparency) that encourages
employees to share their data for workforce analytics purposes.
Project Initiation
Well-managed projects are initiated with clearly defined objectives,
designated project teams with specified roles and responsibilities, defined
deliverables, and an initial high-level timeline. One of the most important
roles is the project manager, who has accountability for fulfilling
commitments, managing the team, communicating with stakeholders,
negotiating changes when needed (for example, extending timelines and
responding to changing priorities), and ensuring high-quality work. During
the initiation phase, the project manager assembles the team, identifies
stakeholders and sponsors, gathers preliminary input and perspectives,
creates a high-level project plan with overall timelines and milestones (see
Figure 14.5), and conducts a project kickoff meeting to discuss and establish
the plans with the project sponsor and team.
Project Execution
When the project is formally kicked off, a detailed project plan might be
necessary. The rigor needed depends on the size and scale of the project. For
more extensive projects (for example, 6 to 12 months or longer in duration),
more detailed project management is needed. This includes specifying tasks,
due dates, owners, dependencies, key meetings, and milestones. The project
manager will track task status on a regular basis, highlight risks, identify and
implement mitigating actions, and escalate issues to senior project leadership
when needed. For shorter projects, a simpler high-level project plan will
suffice. See Figure 14.6 for a sample portion of a detailed project plan.
With project management processes firmly in place, the team can proceed
with the analytics work. This includes further defining the key issues and
hypotheses, identifying data and analysis requirements, collecting and
analyzing data, developing insights and recommended actions, and
developing the story to be communicated.
Project Conclusion
When the project execution phase is complete, closeout actions begin. Design
an implementation plan for the recommended actions. Include key
milestones, implementation responsibilities, change management actions, and
success metrics.
Formal feedback should be given to team members (in addition to ongoing
informal feedback throughout the project duration). Feedback should be
gathered from project sponsors and stakeholders, to get their perspective on
what went well and what future projects can improve.
Finally, the team should discuss and document the lessons learned and the
insights gained from this project. For example, if you learned that you
underestimated the amount of time it would take to obtain data access, build
additional time into your next project plan for that activity.
Accountability
If an organization makes an investment in workforce analytics, it is only
reasonable to expect a positive return on that investment. The analytics team
must demonstrate the value of workforce analytics through a well-structured
business case and the capability to track the team’s performance through a set
of relevant metrics.
“HR has opportunities to think of their employees more like customers and
leverage data that is generated from user interactions. Between 50 percent
and 60 percent of operating expenses are people related. By looking at
employees as customers, there are many business case examples for analytics
In building the business case, you should work with stakeholders and
prospective project sponsors. When consulting with them about what is
possible, you might find it helpful to present case studies of similar work that
has been well executed and clearly shows the return on investment (for
example, quantified value of improved performance, faster time to
productivity, or reduced cost). This helps set expectations and, if necessary,
stimulates conversation about the expected deliverables of the analytics work.
In addition to comprehensively quantifying the benefits of workforce
analytics, you must fully understand the cost of the investment. Specify what
is needed in terms of people (internal staff and external partners), workspace,
tools and technology, and any other expenditure needed. Be sure to obtain the
organization’s commitment to the resources and funding required in
exchange for your commitment to deliver value.
Summary
A well-planned operating model drives efficiencies and allows the workforce
analytics team to focus on the work at hand: successfully implementing
analytics projects that improve business results. Following are the key actions
for establishing your operating model:
• Confirm your vision and mission, incorporating the renewed perspective
that comes with experience.
• Understand data privacy legislation, regulations, and policies for all
countries in which your organization operates, and adhere to them at all
times.
• Understand the sensitivities of HR data; rely on local HR expertise to
******ebook converter DEMO Watermarks*******
guide your analysis and ensure that you operate within policies and
guidelines.
• Request broad permissions for data use when embarking on a new data
collection effort, and encourage employee participation by applying the
FORT framework (gathering feedback, opting in, recognizing
contributions, and ensuring transparency).
• Ensure that the reporting structure provides access to the CHRO.
• Manage your team’s scope and configuration to ensure that analytics does
not get displaced with metrics reporting activities.
• Clearly define the roles and responsibilities of your team and the people
you interact with to execute your mission.
• Agree on a decision-making process that will serve your team well, and
use an advisory panel as part of your process.
• Adopt a consultancy approach to project management, with clearly
defined activities and checkpoints during project initiation, execution, and
conclusion.
• Build a business case at least annually that contains a comprehensive view
of investments and benefits, to demonstrate return on investment and
obtain commitment for funding.
• Agree on and implement your own outcome and proximal metrics to hold
your team accountable for delivering value to the organization.
Perspectives of Analytics in HR
HR professionals have varying degrees of analytics comfort and expertise. A
logical first step in enabling a broad analytics mind-set in HR, therefore, is to
identify where HR professionals fall along the spectrum of analytical
“There are data-savvy individuals and then those who are less numerically
literate. I think that is normal in HR.”
—Andre Obereigner
Senior Manager, Global Workforce Analytics, Groupon
Analytically Savvy
Examples of quantitatively experienced HR professionals abound. These are
often people who joined HR from other functions or were trained in
analytical techniques as part of their formal education. HR professionals in
the analytically savvy category view HR through a business lens, one that
uses analytics to inform decision making and improve performance. They can
be helpful in embedding that mind-set into the culture of HR.
Some HR specialties are more likely than others to have analytically oriented
practitioners; examples include compensation and benefits, employee
engagement, and health and wellness, where numerical literacy is needed for
performing the job. This quantitative focus has helped move the HR function
toward increased financial literacy and has potentially set the stage for
improving analytical literacy.
******ebook converter DEMO Watermarks*******
Finding people in HR who already have analytics capability gives you an
opportunity to leverage the hidden analysts in your organization. These
people, particularly if they are well-respected and influential members of the
HR community, can help their less analytically oriented colleagues
understand and appreciate the value an analytics lens brings to HR. Salvador
Malo, Head of Global Workforce Analytics at Ericsson, says, “Some people
are self-starters and primed for workforce analytics; you should find that
latent skill that exists in organizations. Once you have a few people here and
there, they serve as examples. There are a lot of internal champions.”
Analytically Willing
The analytically willing are open-minded about analytics and prepared to
learn, but they have not had the formal training or exposure needed to
develop analytical skills. Mark Berry says most HR professionals fall within
this category, which he describes as those who “don’t necessarily get it, but
want to understand and use it.” HR training programs and educational
curricula have not traditionally focused on building analytics capability, and
many people who have been attracted to the profession in the past would be
unlikely to name statistical analysis as a particular interest or strength. That
said, members of this group are not opposed to applying analytics to HR;
they simply lack the know-how to do so. This willingness and open-
mindedness presents an encouraging opportunity to build the desired
analytical culture and way of thinking.
“Capability is growing. We’ve had some training efforts to build that, but it’s
really about changing behavior, helping people realize that using data will
empower them and build credibility.”
—Ian Bailie
Global Head of Talent Acquisition and People Planning Operations, Cisco
Analytically Resistant
The analytically resistant group represents the biggest challenge. For a
variety of reasons, some people simply reject the notion that applying an
analytical approach to HR is a valuable and worthwhile endeavor. And if the
analytically resistant are particularly influential members of the organization,
their mind-set, as conveyed through words and actions, can create obstacles
and limit the potential impact of workforce analytics.
The goal in changing this mind-set is not to transform the resistant into
analytics experts, but rather to have them accept and even embrace the
concept that an analytical approach can enhance the value of their business.
Peter Allen, Managing Director, Agoda Outside, stresses the importance of
analytics to HR’s credibility: “HR should be an advisory service that helps
managers to do their jobs well, and to be a credible adviser, you need to be
seen to know what you’re talking about. One of the best ways to do this, of
course, is with good data and meaningful analytics.”
Summary
To reap the most benefit from workforce analytics and achieve the widest
impact, the broader HR function within an organization must embrace and
demonstrate an analytics mind-set through their words and actions. The
following recommendations will help you enable an analytics mind-set in
******ebook converter DEMO Watermarks*******
your HR function:
• Understand the variations of analytics literacy among your organization’s
HR professionals and categorize them accordingly: analytically savvy,
analytically willing, or analytically resistant.
• Customize communications and enablement based on the analytics
literacy categorizations:
• For the analytically savvy, provide training on the latest analytical
techniques, tools, and methods; encourage them to connect with
community forums.
• For the analytically willing, demonstrate and coach them on familiar
data and tools, and have them attend basic statistics courses.
• For the analytically resistant, win them over by using analytics to
provide insights that will make them more successful.
• Identify HR business partners with the right analytical mind-set and skills
to perform the important translator role in workforce analytics.
• Understand your leaders’ views of analytics, leverage their support fully,
and form a coalition of additional supporters as needed.
• Encourage an analytical mind-set in leaders through one-on-one coaching
and involvement in projects that are directly beneficial to them.
Stakeholder Skepticism
Stakeholder skepticism can range from doubts about the value of analytics to
doubts about the value of HR. Following are frequently heard expressions of
doubt and suggestions for addressing them.
Financial Frugality
A strong emphasis on managing costs can sometimes result in missed
opportunities to create future value. This is true for many organizational
decisions, including whether to invest in workforce analytics. Following are
typical financial reasons for holding back and suggestions for addressing
them.
HR Hesitancy
Sometimes the reluctance to embrace workforce analytics comes from the HR
function itself. Potential causes of resistance are many, from not knowing
******ebook converter DEMO Watermarks*******
where to begin, to having concerns about data, tooling, and the ability to
follow through with actions. Following are common concerns from HR and
suggestions for addressing them.
“The advent of cloud technology has brought all this computing power to the
fingertips of people in HR. Now everyone on the planet has access, when
nobody had access before.”
—Josh Bersin
Principal and Founder, Bersin by Deloitte
Table 16.1 Reasons for Analytics Resistance and Tips for Addressing
Them
Warming his hands on a hot drink in a noisy cafe just off 42nd Street
in Manhattan, Jeremy Shapiro, Global Head of Talent Analytics at
Morgan Stanley, talked about what’s new in analytics. The
conversation was not about new methodologies, new technologies, or
even statistical skills. The topic was a critical aspect of analytics,
which also happens to be one of the oldest traditions in human
history: storytelling. Here’s what Jeremy said during that
conversation: “It’s a great feeling when, after your team has worked
so hard to develop strong insights, the story comes through loud and
clear to someone else, particularly to a leader. People can’t make
decisions based on data they don’t understand.”
Given the importance of storytelling in workforce analytics, this
chapter discusses the following:
• Principles and techniques for constructing fact-based stories
• A three-step model and other essentials to build effective
visualizations
• Understanding your audience and tailoring communications
• The need to keep things simple
What Is Storytelling?
People have been telling stories for millennia. The U.S. National Storytelling
Network (www.storynet.org) defines storytelling as “an ancient art form and
a valuable form of human expression.” In her 2002 Harvard Graduate School
******ebook converter DEMO Watermarks*******
of Education article, Deborah Sole describes storytelling as “an ancient and
traditional way of passing on complex, multi-dimensional information and
ideas through narrative.” Nancy Duarte, a TED Talk storyteller and author of
award-winning books on presentation skills, explains that an idea is
powerless if it stays inside you: “Maybe some of you have tried to convey
your idea and it wasn’t adopted; it was rejected and some other mediocre or
average idea was adopted. And the only difference between those two [ideas]
is in the way it was communicated.”
Storytelling is not only about passing on wisdom or sharing ideas; in the
business environment, it is an agent for change. David Boje, a professor at
New Mexico State University, stated in his 2006 Academy of Management
Review article, “Storytelling is widely acknowledged as instrumental to
organization change, training, strategy, and leadership.” Paul Yost, associate
professor at Seattle Pacific University, supported this view in a 2015 paper
when he and his co-authors wrote, “storytelling is becoming an increasingly
utilized art for leading and managing change.”
• Reveal detail and intrigue. Effective stories provide detail that is both
interesting and striking. In George Orwell’s dystopian novel Nineteen
Eighty-Four, the opening paragraph helps draw the reader in with detail
and intrigue: “It was a bright cold day in April, and the clocks were
striking thirteen.” Scriptwriter Andrew Stanton also suggests, in his
“Clues to a Great Story” TED Talk, revealing information progressively.
He describes it as his Unifying Theory of 2 + 2: “Don’t give the audience
four, give them 2 + 2.” The theory is that the story needs to be told
sequentially with emotional elements that unfold to lead the audience
along a journey. In other words, don’t just give your audience the
destination; take them on the journey, too.
Figure 17.4 shows how to add detail to a personalized story.
Although personalization, intrigue, and detail make a story come alive, too
much technical detail can reduce the impact of your story, especially if you
are communicating with nontechnical people. Michael Bazigos, Managing
Director and Global Head of Organizational Analytics & Change Tracking,
Accenture Strategy, explains: “Avoid using the language of statistics, even
though that’s what underlies the analysis.” Patrick Coolen, Manager of HR
Metrics and Analytics at ABN AMRO, agrees: “We always think about how
we’re going to present our insights to each different audience. We leave out
all of the technical information. We work with strong visuals and add
personal details to create the emotional bond.”
Effective Visualization
Having pictures, or visualization, as part of a workforce analytics story is
fundamental to effective communication. Our brains are hardwired to
interpret visual information. As James Zull says in his book, The Art of
Changing The Brain (Stylus Publishing, 2002), we can cognitively process
graphs and images faster than text, which is why visualization is such an
efficient means of communication. However, visualization should not replace
the story; it should be used to enhance the story, draw attention to data, and
more deeply influence some kind of behavior.
Many books and articles delve into visualization with data. Recommended
are any of Edward Tufte’s books, although his first book, The Visual Display
******ebook converter DEMO Watermarks*******
of Quantitative Information (Graphics Press, 1984), is an excellent and often-
quoted text. In addition, the outstanding book Storytelling with Data, by Cole
Nussbaumer Knaflic (Wiley, 2015), is a helpful contemporary reference with
dozens of tips for effective visual aids. This section does not repeat what
those books articulate. Instead, we summarize Edward Tufte’s main elements
for graphical excellence and provide a three-step model to help you create
visual aids in the context of workforce analytics.
Graphical Excellence
Edward Tufte is known for his pioneering thinking and writing on visual
communication of information. His guiding focus is that all visual
communication of quantitative information should be characterized by three
elements: restraint, simplicity, and impartiality.
• Restraint. Data visualization design should not be dictated by how the
visualization can be produced. The design should be conceptualized first,
before creating it with technology. This is especially important today, as
visualization technology claims to automatically present data in new and
exciting ways without the need for any input from analytics practitioners.
Be careful not to rely on visualization technology alone. Take time to
conceptualize and design the visual on paper first, to keep your message
honest.
• Simplicity. Avoid unnecessary visual elements in graphics—what
Edward Tufte calls “chartjunk.” An example of chartjunk is gimmicky
icons or ornate backgrounds on presentation or report pages. The point is
that clean design reflects clear thinking. Do away with unnecessary
aspects that distract the audience.
• Impartiality. In addition to simplicity, visual aids should accurately
represent the data. Avoid distorting the visual chart or graphic with
elements that fail to represent the data properly (for example,
misrepresenting data by skewing the axis of a chart unnecessarily). Such
elements are not only distracting, but they also misguide the viewer.
Eric Mackaluso, Senior Director of People Analytics, Global HR Strategy &
Planning at ADP, provides additional simplicity to Edward Tufte’s advice
with his Four C’s technique for preparing communications materials
(including graphics) for workforce analytics:
******ebook converter DEMO Watermarks*******
• Make it concise.
• Ensure that it looks clean.
• Be sure it is clear.
• Produce it in a captivating way.
As a final point on graphical excellence, Edward Tufte warned in a 2003
article in Wired that PowerPoint is overused as a tool to represent data:
“PowerPoint is a competent slide manager and projector. But rather than
supplementing a presentation, it has become a substitute for it. Such misuse
ignores the most important rule of speaking: Respect your audience.”
You need to decide what medium to use to display the visualization and what
form the visualization will take. By medium, we are referring to whether the
visualization will be viewed on paper or presented on a screen. Furthermore,
if it is screen based, consider whether it will allow a viewer to actively
explore by scrolling, zooming in and zooming out, or drilling in and drilling
out to give greater numerical detail at different levels of analysis (for
example, from organizational-level insights down to divisions and teams).
You might even have multiple visualizations presented in a dashboard, as in
Figure 17.8, which shows levels of employee sentiments and trending topics
amalgamated from social sentiment analysis (adapted from Shami, N.S. et al,
2014).
The data on which the visualization is based might not be static, but instead
gathered from a live data feed. In this situation, you have less opportunity to
control the message you want to convey. If exploration of the data is the top
priority, then dynamic visualizations make sense. However, if the idea is to
convey a specific point, you might prefer static visualizations.
Even with the prevalence of screen-based visualizations, do not discount the
paper option. Admittedly, you have fewer opportunities to explore the data,
and you might even be limited to a black-and-white presentation (as we are in
the print version of this book), but sometimes a simple black-and-white chart
is all you need to make your point.
On the other hand, if you have time series data, you might show changes in
sales performance, for example, plotted as a function of time. Such a
representation would see the time points plotted on the horizontal (x) axis of
a graph and sales performance plotted on the vertical (y) axis of the graph.
Figure 17.10 provides an example of a visualization showing a time series
model of sales performance for five teams. This reveals variability in
performance across the teams but an overall increase in average sales
performance over time.
Note that, with two variables, such as in Figure 17.10, relationships are
relatively easy to visualize. The relationships among three variables can be
visualized with 3D representations, but more than three variables can be
difficult to represent. Analysts often revert to summarizing multiple variables
into just two or three variables using the techniques in Chapter 5, “Basics of
Data Analysis,” or alternatively, presenting the relationships in tables or
correlation matrices.
Keeping It Simple
This chapter has covered a great deal of material about crafting stories,
providing effective visualization, and tailoring your communication to
different audiences, but the final word has to be about the importance of
keeping things simple.
When you have dedicated weeks or even months to detailed data gathering
and analyses, it is tempting to share all that effort with your audience. Don’t.
Laurie Bassi, CEO of McBassi & Company, says: “Do not share everything
you’ve learned—executives don’t care. Less is more.” She always tries to
condense her analytics work to one slide: “No matter how complex the study,
I condense insights down to one slide, with one graph that people can
immediately understand.”
Thomas Rasmussen, Vice President of HR Data & Analytics at Shell, has had
a lot of experience presenting to a variety of audiences. His method for
simplification is based on the practice of writing abstracts: “Think of an
abstract in a journal. Condense your analytics work down to an abstract.
When my team does this, I say, ‘Great, now give me the abstract of the
abstract,’ and keep doing that until the message is succinct and clear.”
Making your presentation simple and getting to the point might take some
******ebook converter DEMO Watermarks*******
time. Many analytics practitioners say that communicating and preparing for
presentations often takes much more time than expected. Whatever time you
allow, it is worth doubling the time you think it will take. As the French
philosopher and mathematician, Blaise Pascal, wrote in the 17th century, “I
have made this letter longer than usual, because I lack the time to make it
short.”
Salvador Malo, Head of Global Workforce Analytics at Ericsson, appreciates
the time it can take to create great communications: “I tell my own team
members to spend as much time on the communication as on the analysis
itself. I can spend time making ten versions or more of the communication
before I deliver it.”
Summary
The art of storytelling and visualization with data requires technique and
practice. Effective storytelling increases the retention rate of the messages
and the likelihood that action will be taken as a result. Visualization helps to
create an understanding of the insights and recommendations, and a good
understanding of your audience helps you tailor communications to maximize
impact. For these reasons, storytelling and visualization are skills that
workforce analytics practitioners should practice. Following are the main
points this chapter emphasizes:
• Ensure that your stories follow these principles: Educate, don’t fabricate;
enlighten, don’t overwhelm; and convince, don’t confuse.
• Create effective workforce analytics stories by providing context, creating
emotional impact, revealing the conflict, and conveying a memorable
message.
• Construct visualizations that show the intent of your analytics by
clarifying the message of your visualization, designing it appropriately,
and testing it before you use it with your selected audience.
• Understand the different types of audiences (newcomer, enthusiast,
scientist, or master) and tailor your communications accordingly.
• Get help from others in preparing your communication materials if you
need it, and strive to keep things simple.
“HR can no longer rely on an old road map to meet the ever-changing
demands it faces. The world of analytics opens up new opportunities that will
enable HR to have different conversations and implement new solutions that
drive better results for the business.”
—Dave Millner
Executive Consulting Partner, IBM
Connected Devices
One major anticipated development is that the data sources that once told us
something about workers at a single point in time (for example, via an
employment survey) will increasingly be replaced by systems that stream
data in real time from technology that is perceptive (sensors), mobile
(wearables), and small (disappearables). Many of these devices are relevant
to workforce analytics and are part of the Internet of Things, an
interconnected world linking electronic devices to others—clocks,
refrigerators, cameras and so on. This means that devices useful for
workforce analytics will be linked to devices used in other areas of the
business, providing greater opportunity for workforce analytics practitioners.
Digital Footprints
Standardized testing of ability or personality scores is common. Our ability to
measure these attributes on job candidates was previously limited to the
people we could encourage to complete a test, either in person or remotely.
Now, however, nontraditional data sources on candidates abound that do not
require the applicants to expend any effort or even know they are being
assessed. Biographical data, or a candidate’s personal history (for example,
educational attainment), can be found, and psychological profiles of
candidates can be derived from digital footprints they leave via their use of
the World Wide Web. Based on this information, businesses can proactively
approach candidates instead of having to wait for the candidates to seek out
opportunities. Digital footprints can be analyzed to help identify change
management influencers. The data can come from sources such as emails,
instant messaging, or social networks.
Genetic Testing
******ebook converter DEMO Watermarks*******
Another new development involves molecular genetic analysis of suitability
for jobs. Recent developments have seen researchers establish links between
genetic profiles and work outcomes. For instance, a Journal of Applied
Psychology article by Chi Wei and colleagues from the National University
of Singapore showed a link between genes associated with extraversion and
job-hopping tendencies.
Importantly, the practical significance of these relationships has not yet been
determined, and the sizes of the relationships so far appear small. Many other
aspects of work behavior are caused by one’s environment, not one’s genetic
profile. Moreover, genetic influences can be switched on or off by one’s
environment. The implications of using these methods need to be better
understood before they are adopted in workforce analytics.
Validity
New data sources come with as many questions as answers about people at
work. The biggest question of all relates to exactly what constitutes
appropriate use. An important factor to consider is whether the methods can
be scientifically demonstrated as effective. For example, although scientific
journals have related social media and text profiling of personality to
standardized personality questionnaires, these relationships are typically
modest: Social media and text-based profiles of personality measure
something similar to standardized questionnaires, but they do not measure the
same concepts. Scientifically speaking, these approaches are likely adequate
for high-volume pre-employment screening, but right now there are better
approaches to use when selecting among the final few candidates.
Legal Appropriateness
Along with considering the scientific merit of utilizing new data sources,
organizations must closely consider the legal appropriateness of doing so.
Legislation generally reacts and follows developments in technology. This is
******ebook converter DEMO Watermarks*******
because the technology that yields the data workforce analytics professionals
might use is evolving quickly and in ways that legislators cannot predict or
imagine. New applications of emerging data sources can be scientifically
validated before they are even considered legal. In other words, science and
technology are often ahead of the law. Deloitte’s 2017 Human Capital Trends
Report explores the rate of change of technology and public policy, including
legislation, in more detail. Different countries also have different laws
regarding the use of technology in workforce analytics. Employment lawyers
should be consulted before new data sources are used in workforce analytics
for employee-related decision making.
Social Impact
Industrial psychologists have developed a thorough understanding of the
social consequences of using different forms of employee-related
information. Using some characteristics in personnel-related decision making
will lead to adverse impact (across socially, legally, or politically salient
groups). However, this understanding is either nonexistent or in its infancy
when it comes to many of the new data sources we have discussed.
Organizations therefore need to carefully consider the social consequences of
the workforce analytics decisions they make based on new data sources. Just
because the science says an application of a new data source is effective and
the law doesn’t say it is inappropriate does not mean that it won’t have
undesirable social consequences.
Evolving Technology
Chapter 11, “Know Your Technology,” outlines the requirements for
establishing an analytics function today. But what technological
developments are on the horizon? This section discusses open standards,
cognitive technology, real-time analytics, and self-service technologies.
Open Standards
The approach of completing time-consuming large technology
implementations before undertaking analytics will fall out of favor as
technology based on open standards becomes the norm. Open standards mean
that HR analytics can begin with the technology on hand today, together with
skilled computer scientists who can integrate the data. We will continue to
******ebook converter DEMO Watermarks*******
see the adoption of cloud technology except where legislation restricts its use.
Artificial intelligence also holds great promise for business and workforce
analytics. Its applications are numerous, for example with bots and digital
agents, where their introduction will change the nature of the workforce, and
hence workforce analytics. This and other emerging technologies bring great
opportunity as discussed in PwC’s 2017 Global Digital IQ® Survey.
Self-service Technology
Expect to see a much wider distribution of analytics capability to workers,
managers, and executives through the use of self-service technology in HR.
This analytical empowerment of workers will become increasingly app
driven and enabled via mobile devices. Numerous benefits follow a shift to
self-service HR environments. First, workers can obtain the information they
need immediately instead of having to wait for a response from HR business
partners. Second, managers will have access to information that allows them
to make insight-driven decisions in real time. Finally, advances in self-service
technology will put information at the fingertips of HR business partners,
enabling them to become more insight driven and strategically influential.
******ebook converter DEMO Watermarks*******
Real-Time Model Updating and Reporting
Practitioners can expect to see much wider adoption in the future of real-time
reporting and model updating. This technology means that when recurring
events happen (for example, employees resign and exit the business at the
end of any particular day), predictive models and reports are updated
automatically and immediately. The latest information can then be passed to
managers using intranet technology, a mobile app, or a text alert. Managers
will be able to take advantage of the outcomes of real-time analytics
immediately instead of having to wait for the results of new models to be
disseminated. These real-time analytics will deliver the information leaders in
organizations need to manage change in increasingly dynamic operating
environments.
Extreme Collaboration
In the future, managing workers will require applying analytics to see the
way HR policies and practices interact with the approaches being applied in
other functions, such as finance, marketing, and legal departments. The new
approach will view HR practices and processes as just one factor that impacts
the way organizations turn inputs (for example, materials and labor) into
outputs (that is, products and services) for customers.
The workforce analytics function therefore needs to embrace the approaches
and language of other functions to fully understand the likely complex ways
that HR policies and practices impact business performance. These demands
will see the workforce analytics function work more collaboratively, and with
a broader range of colleagues and stakeholders, than it has to date. Expect to
see a greater number of partnerships as projects become more complex and
require abilities beyond the immediate capabilities of the team. This
integration will also see the data that once flowed only within silos in
businesses become available in real time for authorized users across the
organization.
Maturity Models
Perhaps notable due to its absence in this book is the notion of analytics
maturity models. Maturity models in analytics essentially specify a roadmap
that organizations can choose to follow to develop their workforce analytics
capability. In the early days of workforce analytics, maturity models were
useful because they highlighted the possibilities for workforce analytics at a
time when the discipline was finding its feet. They spotlighted the usefulness
of data warehouses (the reporting level in many frameworks) and the
potential to predict employee events with a useful degree of accuracy (the
predictive level in many maturity frameworks).
******ebook converter DEMO Watermarks*******
In the future, the implied need to follow a linear progression through
seemingly more sophisticated levels of analytics capability is a
misrepresentation of what is needed and what is possible for most
organizations. In fact, all forms of workforce analytics are likely to be needed
concurrently. Organizations do not need to surmount the hurdles associated
with earlier levels of an analytics maturity framework before reaching the
desired stage of workforce analytics capability. Instead, organizations should
simply focus on the analytics capability they require to solve their most
pressing business challenges. In other words, don’t focus on the maturity
model which is an inside-out view of workforce analytics. Instead focus on
the business issues and the stakeholders that are served, which gives an
outside-in emphasis.
Summary
Workforce analytics is an essential organizational capability and experts
believe its impact on business performance will only increase. By taking the
following steps, you can position your function well to capitalize on the
opportunities:
• Familiarize yourself with emerging data sources in workforce analytics,
such as worker digital footprints, sensors, and wearable data.
• Ensure that your function is ready to utilize new and emerging technology
that blends reporting and predictive analytics using open source
technology, all deployed with artificial intelligence and cognitive
technology.
• Decide on the appropriateness of data sources for particular purposes
based on the scientific evidence, legal context, and social implications of
using the data.
• Prepare your team to evolve its skills, particularly the ability to lead in a
complex environment, with privacy as a specialty and using data scientists
who can learn new methods and techniques quickly.
• Learn why workforce analytics is important for tomorrow’s business and
prepare to embrace the road ahead.
B
A/B testing, 154
Bailie, Ian, 9, 12, 181, 231, 233, 243, 246
Bassi, Laurie, 93, 104, 105, 115, 138, 194, 273
C
Callery, John, 89–90
capability, as driver of employee engagement, 65–66
CAPEX (capital expenditure), 165
case studies
D
data analysis. See also data quality
correlational designs, 46
data governance, 152–153
experimental design, 45
experimental designs, 45
longitudinal designs, 47
objectives of, 47–48
overview of, 43–44
qualitative analysis, 51–52
qualitative studies, 47
quantitative analysis, 48–51
quasi-experimental designs, 45–46
research design, 44
social consequences of, 54–56
traditional statistics versus machine learning, 53–54
unstructured data, 52–53
data awareness, 179
data complexity, 130
data dictionaries, 153
data gathering, 31–32
data governance, 152–153
data outliers, 145–146
data owners, 107
data privacy, 176–177
data quality
A/B testing, 154
Big Data, 152–155
E
Ebelle-ebanda, Antony, 91, 241, 271
Edwards, Martin, 56
effective visualization, 263
efficiency, operational, 93
Eight Step Model for Purposeful Analytics, 28, 58
emotional attachment, creating, 257–259
employee engagement. See engagement
Employee Net Promoter Scores (eNPS), 66
enabling analytical thinking. See analytical thinking, enabling
engagement
business impact of, 65
ISS Group case study
F
FA (factor analysis), 50
failure of analytics, reasons for, 40–42
fairness–aware data mining, 54–55
feedback, 207
Feffer, Mark, 9
Ferris, Gerald, 129–130, 171
finance department, engaging, 110–111
financial frugality, overcoming, 245–246, 251
financial literacy, 171
Fink, Alexis, 22, 25, 37
focus of workforce analytics, 16–17
FORT framework, 207
framing business questions, 29–30
Frampton, Jonathon, 37, 273
Friedman, Jerome, 56
frugality, overcoming, 245–246, 251
funding, 110
G
gathering data, 31–32
genetic testing, 280
Getinge Group, 43, 192, 261
gig economy, 12, 18
GlaxoSmithKline, 36
Global Consortium to Reimagine HR, Employment Alternatives, Talent, and
the Enterprise (CHREATE), 11, 18
******ebook converter DEMO Watermarks*******
“good enough” data, 138–139
governance, 152–153
operating model
decision-making approach, 210–212
HR data, working with, 205–208
reporting structure optimization, 208–209
Rentokil Initial case study, 71
graphics, 263–264
Green, David, 57
Groupon, 102, 147, 228, 231, 248
H
Handbook of Industrial and Organizational Psychology (Lee), 52
Hartmann, Peter, 43, 192, 261
Hastie, Trevor, 56
health and wellness, requests from managers, 8
Herleman, Hailey, 160
hesitancy, overcoming, 246–251
HIPPO principle, 100
in–house teams, 189–191, 196
HR (human resources)
changing nature of, 7
consumerization of, 9
contribution to business value, 4–6
data warehouses, 160–161
democratization of, 7–9
future of, 10
hesitancy, overcoming, 246–251
HR leaders, engaging, 102–103
lack of skills in, 242
perception of, 199–200
requests from managers, 8–9
skills of, 175–177
I
IaaS (Infrastructure as a Service), 165
IBM, 11, 52, 57, 112, 150, 157, 278
impact, 55
impact assessment, 132–133
impartiality, 264
impasse, resolving, 211
implementation, 36–38
implementation complexity, 131
operating model
consulting approach to project management, 217–221
role/responsibility clarification, 214–216
team structure, 213–214
of recommendations, 119
risks, 249–250
inconsistent data definitions, 146–147
industrial psychology, 177–178
Infrastructure as a Service (IaaS), 165
J
Jover, Placid, 110, 214, 220
JPMorgan Chase & Co. 89, 92, 119, 231, 250
K
Karia, Akash, 256
key performance indicators (KPIs), 6, 125
Klavohn, Travis, 16
Klinghoffer, Dawn, 177, 284
Knaflic, Cole Nussbaumer, 263
KPIs (key performance indicators), 6, 125
kurtosis, 143–145
L
lack of data, 142–143, 249
Landstinget Västmanland, 239
******ebook converter DEMO Watermarks*******
Lashyn, Terry, 22, 36, 42, 175
Lawler, Ed, 16
Layney, Tracy, 6, 227
leadership
attributes of, 23–26
board of directors, 105
building analytics culture with, 236
business acumen of, 22–23
business leaders, 103–105
chain of command, 20
data owners, 107
HR leaders, 102–103
requests from managers, 8
responsibilities of, 20–21
SMEs (subject matter experts), 108–109
technology owners, 108
leasing technology, 165–166
Lee, Tomas, 52
legal appropriateness of data sources, 281
legal department, engaging, 109–110
Levenson, Alec, 5, 28–29
Lever Analytics, 137, 194
linguistic analysis, 280
LinkedIn, 172, 203, 208
listening
analytics team responsibilities, 118–120
to project sponsors, 90–91
London Metropolitan Police. See Metropolitan Police case study
longitudinal designs, 47
M
machine learning, 53–54, 162–163
Mackaluso, Eric, 206, 243, 264
N
Nagy, Mihaly, 113, 200
National Storytelling Network, 254
negotiating scope of analytics, 95
Nestlé, 35, 272, 277
new data sources, 277–280
evolving technology, 282–284
issues surrounding, 280–281
newcomer audiences, 271
Nicholas, Ben, 36
Nielsen, Adam Chini, 116
Nielsen Holdings PLC case study, 59–63
analysis of attrition, 60–62
O
Obereigner, Andre, 102, 147, 228, 231, 236, 248
Oest, Martin, 76, 111
O’Hanlon, Peter, 137, 194
O’Keefe, Ian, 88, 92, 119, 231, 250
O’Neill, Cathy, 54
on–premise technology, 164
open standards, 282
operating model
accountability
business cases, 221–222
success metrics, 222–223
chain of command, 20
defining, 204
governance
decision-making approach, 210–212
HR data, working with, 205–208
reporting structure optimization, 208–209
implementation
consulting approach to project management, 217–221
role/responsibility clarification, 214–216
team structure, 213–214
linking to strategy, 205
operational efficiency, 92
opportunity cost, 133
optimizing reporting structure, 208–209
P
PaaS (Platform as a Service), 165
partners
advantages of, 188
budget considerations, 198
factors in selecting, 197–200
outsourcing, 193–197
in–sourcing, 191–193, 196
Parveen, Sofia, 116
Pascal, Blaise, 273
PCA (principal components analysis), 50
Pearson’s product–moment correlation, 49
people analytics. See workforce analytics
perception of HR, 199–200
personalization, workforce, 9–10
perspectives
analytically resistant professionals, 232–234
analytically savvy professionals, 228–230
analytically willing professionals, 230–232
Pet Projects, 128
Q
qualitative analysis, 51–52
qualitative studies, 46–47
quality. See data quality
quantitative analysis, 48–51
quantitative skills, 178–179
quasi–experimental designs, 45–46
Quick Win projects
complexity–impact matrix, 125–126
definition of, 123–124, 126–127
identifying, 124–125
R
RACI responsibility matrix, 214–216
Rasmussen, Thomas, 92, 99, 102, 104, 118, 191, 213, 229, 231, 273
real-time model updating, 283
recognition, 207
recommendations
determining, 33–34
implementation of, 119
recruitment, 8
reduction analysis, 50
reduction of costs, pressure for, 93
******ebook converter DEMO Watermarks*******
regression analysis, 62
regulatory requirements, 92
relationship power, 40
Rentokil Initial case study, 71–75
implementation of recommendations, 74–75
iterative approach, 72–73
overview of, 71
variability in sales performance, 71
reporting, 17
real–time reporting, 283
reporting structure, optimizing, 208–209
technology, 161–162
representation, responsibility for, 21
requests from managers, 8–9, 92–93
research design, 44, 178
resistance, overcoming, 239–240
financial frugality, 245–246, 251
HR hesitancy, 246–251
stakeholder skepticism, 240–245, 250–251
responsibility matrix, 214–216
responsibility/role clarification, 214–216
restraint, 263
retention analysis, Nielsen Holdings PLC case study
analysis of attrition, 60–62
impact of increasing attrition, 59–60
implementation of recommendations, 63
overview of, 59
return on investment (ROI), 37, 132
revealing conflict, 259–260
revealing insights, 33–34
robotics, 18
ROI (return on investment), 37, 132
roles (team), 181–185
S
S&P Global, 91, 241
SaaS (Software as a Service), 164–165
Salapatas, Kanella, 159, 192, 271
sales growth, Rentokil Initial case study
implementation of recommendations, 74–75
iterative approach, 72–73
overview of, 71
variability in sales performance, 71
scene, setting, 256–257
scientist audiences, 270
scope, agreeing on, 95
segmentation analysis, 50
self-service technology, 282–283
Seligman, Martin, 93
sensitive personal information (SPI), 205
setting the scene, 256–257
Seven Forces of Demand, 91–95
Shami, Sadat, 52, 112
Shapiro, Jeremy, 152, 187, 242, 253
sharing technology, 165
Shell, 92, 99, 102, 104, 118, 191, 213, 229, 273
Shutterfly, Inc., 6, 227
similarities, emphasizing, 268–269
simplicity, 263, 273–274
Six Skills for Success
business acumen, 22–23, 170–172
T
talent, definition of, 16
talent analytics. See workforce analytics
Talent Strategy Institute, 102, 113, 183, 272
talent supply chain, 181
Tanfeeth, 153, 271
teams, 188–189
development, 24
in-house, 189–191, 196
outsourcing, 193–197
partners
advantages of, 188
budget considerations, 198
factors in selecting, 197–200
outsourcing, 193–197
in–sourcing, 191–193, 196
role/responsibility clarification, 214–216
Six Skills for Success
business acumen, 170–172
communication, 179–181
U
Unilever, 110, 214, 220
unions, engaging, 112
unstructured data, 52–53
U.S. National Storytelling Network, 254
V
validity of data sources, 280–281
vendor relationships, 165–167
vignettes
Allen, Peter, Agoda.com, 94
Andersen, Morten Kamp, proacteur, 40
Bailie, Ian, Cisco, 181
Berry, Mark, CGB Enterprises, 262
Callery, John, BNY Mellon, 89
Champ, Marcus, Standard Chartered Bank, 127
Coolen, Patrick, ABN AMRO, 195
Dellala, Damien, Westpac Group, 209
******ebook converter DEMO Watermarks*******
Dillon, Sally, ANZ Bank, 162
Duin, Eric van, PostNL, 124
Everduin, Giovanni, Tanfeeth, 153
Fink, Alexis, Intel, 25
Huselid, Mark, Northeastern University, 10
Jover, Placid, Unilever, 220
Klinghoffer, Dawn, Microsoft, 177
Layney, Tracy, Shutterfly, Inc., 6
Malo, Salvador, Ericsson, 232
Nielsen, Adam Chini, Nordea, 116
Obereigner, Andre, Groupon, 248
Oest, Martin, True Picture Europe Ltd., 111
Parveen, Sofia, Nordea, 116
Rasmussen, Thomas, Royal Dutch Shell, 191
Salapatas, Kanella, ANZ Bank, 159
Smeyers, Luk, iNostix by Deloitte, 244
Sonnenberg, Mariëlle, Wolters Kluwer, 148
Voorn, Bart, Ahold Delhaize, 235
White, Rebecca, LinkedIn, 172
Wright, Patrick, University of South Carolina, 106
Yost, Paul, Seattle Pacific University, 274
vision
developing, 95–97
operational model and, 205
technology and, 158–159
The Visual Display of Quantitative Information (Tufte), 263
visualization
creating, 264–269
effective visualization, 263
graphics, 263–264
technology, 163, 180
testing, 269
Voorn, Bart, 32–33, 93, 235
X-Y-Z
Yost, Paul, 254, 260, 274
Zak, Paul, 254
Zull, James, 263