1/18/2016

HYPOTHESIS-DRIVEN
ENTREPRENEURSHIP
ENGR 112: Laboratory to Market, Entrepreneurship for Engineers
Winter 2016, Lecture 4
Nathan M. Wilson, Ph.D., M.B.A.
Anderson School of Management
Department of Mechanical and Aerospace Engineering, Henry Samueli School of Engineering and
Applied Sciences
Schaffer Grimm, M.S., M.B.A.
Lecturer, Henry Samueli School of Engineering and Applied Sciences
Manager of Strategic Business Planning, Institute for Technology Advancement

Syllabus: Lectures
Week

Session Description

Introduction to Entrepreneurship; Small Business Paths

Business Model Canvas

Hypothesis-Driven Entrepreneurship; Customer Discovery

Markets & Industries; Industry Analysis; Marketing

Distribution Channels; Partners; Competitors

Revenues & Costs; Entrepreneurial Accounting

Mid-Term Exam & Entrepreneurship in Practice

Legal Issues and IP

Raising Capital

Introduction to Business Plans & Elevator Pitches;

10

Disruptive Innovation and Technology Transfer

11

Final Presentations and Final Exam

1/18/2016

Agenda
Motivation for hypothesis-driven entrepreneurship (the Lean

Startup method)
Hypothesis-driven entrepreneurship (HBS)
Test (VPD)

Three Other Common Approaches to Launching a Business

Build-It-And-They-Will-Come
Relies on a founders vision; engineering dominated team delivers product

Waterfall Planning
Product development divided into phase; sequentially completed by different

organizational units

Just Do It!
Uses an improvisational approach; adapting to feedback from resource

providers and customers.

1/18/2016

Three Fallacies of Business Plans

1.

They rarely survive first contact with customers.

Everyone has a plan until they get punched in the mouth. Mike Tyson

2.

Five-year plans are generally fiction

Forecasting complete unknowns is almost always a waste of time.

3.

Start-ups are not smaller versions of large companies.

Existing companies execute a business model; start-ups look for one.

Decades-Old Formula vs. Lean Startup

TRADITIONAL

LEAN STARTUP

Write a business plan

Test hypotheses

Pitch it to investors

Gather early & frequent customer

Assemble a team
Operate in stealth mode
Release fully functional prototype

75% of all start-ups fail

feedback
Show MVP
Pivot

Refining along the way for a

higher likelihood of success

1/18/2016

Business Plans vs. Experimentation Process

Instead
1.

Summarize hypotheses in a business model canvas

Diagram of how a company creates value for itself and its customers.

2.

Get out of the building approach

Rapidly assemble MVPs and ask immediately ask for feedback.

3.

Practice agile development

Develop the product iteratively and incrementally

1/18/2016

In the Past: Five Factors Constraining Startups

High cost of first customer and the even higher cost of getting

the product wrong.

Long technology development cycles.
Limited number of people with propensity for risk inherent in

founding or working at a start-up.

The structure of the VC industry
(A few funding firms and a large number of start-ups looking for funding.)

Concentration of expertise in building startups

(In US, pockets on the east and west coasts.)

10

Claim: Lean Methods = Fewer Failures

How they are positioned to succeed:
Lean start-ups launch products customers want
Early feedback, iterative versions of product

Lean start-ups launch products more quickly and more cheaply

Open source software, manufacturing in China, etc.

Decentralization of access to financing

Now there are super angel funds, accelerators, crowdfunding, etc.

Availability of information
The Internet offers so much data, the biggest challenge is sorting through all of it.

1/18/2016

11

I-Corps Team Lessons Learned Videos (LA Node, Fall 2014)

Team 414
http://www.youtube.com/watch?v=9RabmgNRYOk
Team 408
http://www.youtube.com/watch?v=vzDCIl5728w

12

10 Testing Principles

1/18/2016

13

10 Testing Principles (cont.)

9 Guesses

Guess

Guess

Guess

Guess
Guess

Guess

Guess

Guess

Guess

1/18/2016

2012 Steve Blank

2012 Steve Blank

1/18/2016

2012 Steve Blank

2012 Steve Blank

1/18/2016

Customer Development

20

Stage Goals
Customer Discovery
Prove the product solves a problem for an identifiable group of users

Customer Validation
Prove the market is salable and large enough that a viable business can
be built
Company Creation
Business is scalable through a repeatable sales and marketing roadmap
Company Building
Company departments and operational processes are created to support
scale

10

1/18/2016

21

Customer Discovery Goals

Validate Problem/Solution Fit
Ensure that you are trying to solve a problem that is important enough
for somebody to pay you to solve
Minimum Viable Product
Propose a product with the fewest number of features needed to solve
the problem, that users are willing to pay for
Propose Sales / Marketing Roadmap
Propose a roadmap for the sales and marketing team that can generate
demand for your product

22

Goals of Customer Validation

Goal to create a repeatable sales road map for the sales and

marketing teams that will follow later

Product-Market Fit
when a product shows strong demand by users representing a sizable
market
Business Model
The cash is flowing in the right directions from the right people to sustain
the business
Sales and Marketing Roadmap
There is a predictable and repeatable process for creating customer
demand and closing sales

11

1/18/2016

23

Product-Market Fit
Customer is willing to pay for the product
2. The cost of acquiring the customer is less than what they pay
for the product.
3. Theres evidence indicating the market is large enough to
support the business.
1.

24

Hypothesis-Driven Entrepreneurship / Lean Startup

Since most startups have tremendous uncertainty about their

proposed business models viability, the hypothesis-driven

entrepreneurship approach helps to maximize feedback for
resolving this uncertainty.
Lean startup: Organizations that follow the principles of
hypothesis-driven entrepreneurship. [Eric Reis]
Lean does not mean that the company is bootstrapping,
but rather that they are avoiding waste (similar to lean
manufacturing).

12

1/18/2016

25

The Hypothesis-Driven Approach

Its all about fast & frugal tests, since time is often the scarcest resource.
Step 1: Develop a vision
Step 2: Translate vision into hypotheses

Step 3: Specify MVP (minimum viable products) tests

Step 4: Prioritize tests
Step 5: Learn from MVP tests
Step 6: Persevere, pivot, or perish
Step 7: Scaling and Ongoing Optimization
Repeat until all you can validate all key business model hypotheses.

26

Lean Startup
Flow diagram

13

1/18/2016

27

Step 1: Developing a Vision (with Ideation)

Immersion
Deep dive
Obsession
Obsessed with the problem, not the solution
Incubation
Subconscious engagement
Recombination
Expose yourself to diverse ideas
Clarification
Keep track of and refine ideas
Collaboration
Cofounders support each other, trigger ideas

28

Step 2: Translate the Vision into Hypotheses

Translates vision into falsifiable business model hypotheses.
if the plan is to see what happens, a team is guaranteed to succeed at
seeing what happens but wont necessarily gain validated learning.
For each business model element, formulate a set of falsifiable

hypotheses.
Use hypotheses that have quantitative metrics when possible
Cohort Analysis: track trends for customers acquired during a
specific period of time (typically with the same marketing
method)
A/B Testing: divide a set of similar prospects or customers into
a control group that experiences a status quo product and a
treatment group that experiences a product with at least one
modified element.

14

1/18/2016

29

Step 3: Specify MVP Tests

Minimum Viable Product (MVP): The smallest set of features

and/or activities needed to complete a Build-Measure-Lean

cycle and thereby test a business model hypothesis.
MVPs may constrain product functionality and/or operational
capability.
Constrained product functionality: only a subset of the features

envisioned
Operational capability: rely on a temporary or makeshift technology to

deliver the MVPs functionality.

Smoke Tests: a product that is truly minimal because it does not yet exist.

Benefits of MVPs:
Shorten product development cycle accelerating customer feedback
Release feature revisions in small batches making it easier to interpret
test results and to diagnosis problems

30

Concerns about MVPs

Exposure to Idea Theft
Ideas are typically worthless unless executed

Reputational Risk
MVP testing limits the target customer base to minimal scale necessary,
thereby minimizing risk
Can potentially use an alternative brand name for the MVP

15

1/18/2016

31

Step 4: Prioritize Tests & Step 5: Learn from MVP Tests

Usually give priority to tests that can eliminate considerable

risk at a low cost

Parallel Testing: potentially can make sense to purse tests in
parallel because the relevant hypotheses are not serially
dependent.
In a winner-take-all-market, parallel testing may offer benefits

Potential issues when learning from MVPs:

False positives & false negatives
Sometimes customers stated preferences dont match their true
preferences
Sometimes entrepreneurs see what they want to see, and they see what
they expect to see

32

Step 6: Persevere, Pivot, or Perish

Persevere
If the MVP test validates a business model hypothesis, test remaining
hypotheses or prepare to scale.
Pivot
If the MVP test invalidates an aspect of the business model hypothesis,
change some business model elements while retaining others .
Perish
If an MVP test decisively rejects a crucial business model hypothesis, and
there isnt a plausible pivot, shut down the business.

16

1/18/2016

33

Step 7: Scaling and Ongoing Optimization

If all key business model hypotheses have been validated, then

a product-market fit has been achieved.

Product-Market Fit: the right product for the market: one with
demonstrated demand from early adopters and with solid
profit potential.
Continue to utilize hypothesis-testing methods, but now for
optimization instead of validation.

34

Introducing the Customer Development Process

17

1/18/2016

35

Integrating Lean Startup Principles: Build-Measure-Learn

36

Apply Build, Measure, Learn

Conceptual Prototypes
Design rapid conceptual prototypes to shape your ideas, figure out what
could work, and identify which hypotheses need to be true to succeed.
Hypotheses
Design and build experiments to test the hypotheses that need to be true
for your idea to succeed.
Start with the most critical hypotheses that could kill your idea.
Products and Services
Build MVPs to test your value propositions.
MVPs used to learn rather than sell.
MVPs contain minimum feature set designed to learn.

18

1/18/2016

37

Testing the Circle (Customers)

Provide evidence showing what customers care about before

focusing on how to help them.

38

Testing the Square (Value Proposition)

Provide evidence showing that your customers care about how

your products and services kill pains and create gains.

19

1/18/2016

39

Testing the Rectangle (Business Model)

Provide evidence showing that the way you intend to create,

deliver, and capture value is likely to work.

40

Overview of the Testing Process

20

1/18/2016

41

10 Testing Principles
Make it visual and tangible
Embrace a beginners mind
Dont fall in love with first ideas create alternatives
Fell comfortable in a liquid state
Start with low fidelity, iterate, and refine
Expose your work early seek criticism
Learn faster by failing early, often, and cheaply
Use creativity techniques
Create Shrek models
Track learnings, insights, and progress

42

Extract Your Hypotheses

What needs to be true for your hypothesis to work?
Use the VPC and BMC to identify what to test before you get out of the
building
Define the most important things that must be true for your idea to work
Business Hypothesis: Something that needs to be true for your idea to

work partially or fully but that hasnt been validated yet.

21

1/18/2016

43

Prioritize Your Hypotheses

What could kill your business?
Not all hypotheses are equally critical.
Start prioritizing whats critical to survival.

44

Design Your Experiments with the Test Card

22

1/18/2016

45

Capture Your Insights with Learning Card

Invalidated
Get back to the drawing board
(i.e. pivot).

Learn More
Seek confirmation
Deepen your understanding
Validated
Expand to next building block
Execute

46

How Quickly Are You Learning?

23

1/18/2016

47

Five Data Traps to Avoid

48

Five Data Traps to Avoid

False-Positive Trap
Seeing things that are not there.

False-Negative Trap
Not seeing things that are there.
The Local Maximum Trap
Missing out on the real potential.
The Exhausted Maximum Trap
Overlooking limitations (e.g. of a market).
The Wrong Data Trap
Risk: searching in the wrong place.

24

1/18/2016

49

The Testing Process

50

Six Techniques to Gain Customer Insights

25

1/18/2016

51

The Data Detective

Build on existing work with (desk) research. Secondary research
reports and customer data you might already have provide a
great foundation for getting started. Look also at data outside of
you industry and study analogs, opposites, or adjacencies.

Difficulty level: *
Strength: great foundation for further research.
Weakness: static data from a different context

52

The Journalist
Talk to (potential) customers as an easy to gain customer
insights. Its a well-established practice. However, customers
might tell you one thing in an interview but behave differently in
the real world.
Difficulty level: **
Strength: quick and cheap to get started with first learnings and
insights.
Weakness: customers didnt always know what they
want and actual behavior differs from
interview answers.

26

1/18/2016

53

The Anthropologists
Observer (potential) customers in the real world to get good
insights into how they really behave. Study which job they focus
on and how they get them done. Note which pains upset them
and which gain they aim to achieve.

Difficulty level: ***

Strength: data provide unbiased view and allow
discovering real world behavior.
Weakness: difficult to gain customer insights
related to new ideas.

54

The Impersonator
Be your customer and actively use products and services.
Spend a day or more in your customers shoes. Draw from your
experience as an (unsatisfied) customer.
Difficulty level: **
Strength: firsthand experience of jobs. Pains, and gains.
Weakness: not always representative of your real customer or
possible to apply.

27

1/18/2016

55

The Co-creator
Integrate customers into the process of value creation to learn
with them. Work with customers to explore and develop new
ideas.
Difficulty level: *****
Strength: the proximity with customers help
you gain deep insights.
Weakness: may not be generalized to all
customers and segments.

56

The Scientists
Get customers to participate (knowingly or unknowingly) in an
experiment. Learn from the outcome.
Difficulty level: ****
Strength: provides fact-based insights on real-world behavior;
works particularly well for new ideas.
Weakness: Can be hard to apply in existing
organizations because of strict (customer)
policies and guidelines.

28

1/18/2016

57

Produce Evidence with a Call to Action

Use experiments to test if customers are interested, what

preferences they have, and if they are willing to pay for what
you have to offer.
Call to Action (CTA)
Prompts a subject to perform an action

Used in experiment in order to test one or more hypotheses

Use experiments to test:

Interest and relevance
Priorities and preferences
Willingness to pay

58

Experiment Library
Ad Tracking
Unique Link Tracking

MVP Catalog
Illustrations, Storyboards, and Scenarios
Life-Size Experiments
Landing Page
Split Testing
Innovation Games
Speed Boat
Product Box
Buy a Feature
Mock Sales
Presales

29

1/18/2016

59

The Progress Board

60

The Progress Board

30

1/18/2016

61

Limits to Lean Startup Methods

The lean startup approach is not recommended in these situations:
When mistakes must be limited
When uncertainty about customer demand is low
When long product development cycles preclude launching early and often
When resources are plentiful

Instead, these companies should employ modified lean techniques or

an alternate development plan.

62

Just Old Wine in a New Bottle?

The components of hypothesis-driven entrepreneurship are

not new
But this approach focuses less on the product and more on the

business model
Balances the strong direction of the founder
with redirection from market feedback.

31