Validating Your Startup Idea

VALIDATING YOUR STARTUP IDEA
NOVEMBER 2020
IDEATION AND VALIDATION
Ideate Get inspired and come up with ideas

Ideation
Litmus test See which ideas are worth pursuing further
Identify potential business models, create

Identify key hypotheses strawman unit economics, and develop key
Validation
hypotheses to test validity of models
Try to prove or disprove each statement,

Test hypotheses starting with what assumptions would “break”
the model if disproven
It’s critical to:

● Quickly pivot or kill initial ideas that don’t meet immediate criteria
● Adjust assumptions and iterate on testing methods as you go along

Ideation

Validation

It’s critical to:

WHAT IS A WELL-DEFINED HYPOTHESIS?
HYPOTHESES SHOULD MEET KEY CRITERIA:
Category Key question Key criteria
Is this a big enough bottoms-up Big enough to build $1b+ company (conservatively assuming with a
Market Size
opportunity? small share of pie and a small take rate)
Do you have competitive No major direct competitors (if winner-take-all situation)

Competition
differentiation & sustainable moat? Long-term moat to protect against new entrants (network effects)
High fragmentation in market
Fragmentation How fragmented is the market?
Scalable tech enablement to aggregate & vet / manage quality
Does this meet acute customer pain Clear value prop for supply
Value Prop
points? Clear value prop for demand
Business model with clear monetization opportunities
Can you get to strong unit
Unit Economics High AOV and/or high frequency of purchases
economics?
Realistic path to achieve net LTV/CAC @12mo ≥2x
Can you build scalable acquisition Repeatable & scalable acquisition engine
Acquisition
channels? If B2B <1 yr sales cycles
BUILD UNIT ECONOMICS STRAWMAN
Framework Understand critical levers
AOV (GMV per order) • In your model, understand the most

Take rate % sensitive inputs that would impact
= Revenue per order
LTV/CAC, and by what degree of magnitude
(-) COGS (e.g., Shipping, packaging) • For critical levers, identify what customer,
= CM1 (Gross Margin) competitor, and market behaviors impact
those in a tangible way
Non COGS variable costs: • Set “low vs. high scenarios” to understand
(-) Landing costs: logistics, packaging, handling, fulfillment lower and upper bounds, and play around
(-) Customer service (e.g. call center agents or reps) with assumptions
(-) Payment: (e.g. credit card fees)
• Record your rationale for every input (why
= CM2
you made that assumption, metrics comps,
etc.)
Paid CAC:
(-) Offline marketing costs
(-) discounts , coupons
= CM3
(-) Fixed costs (e.g. salaries, rent, other overhead)
= EBITDA

Ideation

Validation

It’s critical to:

THERE ARE 2 TYPES OF RESEARCH
Secondary Research Primary Research
● Conducted by others ● Self-conducted

● Typically quick to find ● Can require a budget and turnaround
● Can be free or paid time
What ● This is where most research begins ● Due to its nature, usually takes more
time to collect than secondary research
● Generally fills the gaps not answered by
secondary research
● S-1s, 10-Ks, etc ● Conversations, interviews

● Equity research papers & analyst ● Surveys
Examples reports ● SEM analyses
● Open APIs & public data sets ● Performance marketing tests
● Customer help & support pages ● Outbound email tests
SECONDARY RESEARCH: RESOURCES
HIGH-LEVEL MORE GRANULAR
• GDP & spend behavior: IBIS, Statista, • S-1s, 10-Ks, 10-Qs, investor presentations for public
Nielsen, Bureau of Economic Analysis, companies can provide helpful macro information and
Bureau of Labor Statistics, Permanent details where players is heading
Equity’s ‘Open Data Resources’ database • Equity research papers & analyst reports from various
• Competitive insights & data: firms (e.g., Credit Suisse, Accenture)
Crunchbase, TechCrunch, Pitchbook, The • Open APIs & public data sets can offer rich information
Information, MarketWatch, Wall Street and API documentation can provide more insight into a
Journal tech player’s products and metrics
• Industry publications: vertical-specific
• Customer help & support pages offer helpful info
newsletters and blogs, podcasts,
including UX screenshots and onboarding flow (e.g.,
community forums, Reddit discussions,
company website FAQs, public Zendesk sites, and other
Twitter
support pages)
PRIMARY RESEARCH: RESOURCES
3 MAIN CATEGORIES
• Customer Discovery: conversations, interviews, and surveys

• Sales & Marketing Tests: SEM analyses, performance marketing tests,
outbound email tests
• Data Collection: scale granularity of data collection and research via
Upwork freelancers
CUSTOMER DISCOVERY
INTERVIEWS & SURVEYS

Example outputs:
Test and refine value props
• Talk to 30 potential customers (network, cold outreach)
● Ecosystem map of key players in end-to-
• Send out online surveys (if appropriate) to quantify sentiment end value chain, including their roles and
and stack rank value props incentives
Deeper understanding of decision-making ● Process map of customer buying journey,
• Purchasing: How do customers approach this today? What are outlining critical pain points and opportunities
their current alternatives? What do they think of them? How do for tech enablement
they think about willingness to pay? ● List of key purchasing criteria for new
• Retention: Why do customers stay? Why have they churned in buyer; list of key expectations for existing
the past? What drives more frequent purchases? customer
As you go along… ● Refined numerical assumptions around
historical transactions (AOV and frequency),
• Understand all players and their motivations in the end-to-end
sales % conversion funnel, etc, to build more
value chain (buyers, suppliers, partners)
robust bottoms-up TAM and unit economics
• Ask yourself: what business model can deliver on these key
● Initial leadgen lists for suppliers or
value props in scalable, sustainable, and differentiated way?
customers (important for B2B to understand
• Refine and adapt how you ask questions to get the best and role / companies to target)
most relevant insights (check out ‘The Mom Test’)
SALES & MARKETING TESTS
PERFORMANCE MARKETING & OUTBOUND TESTS

Example outputs:
Analysis, then paid tests

● Performance marketing budgets
● Get an idea on traffic and visibility from organic and paid based on key inputs: Cost-per-click
search (CPC), Conversion rates, estimated
● Narrow down target demographic and popular categories CAC
(specific keywords and audience criteria) from volume
● Competitor acquisition channels
analysis
based on assumptions from SEO (can
B2B nuances leverage Spyfu)
● Lower volume and longer sales cycles ● Category and demo volumes,
● B2B testing does not need to be statistically significant leveraging metrics such as: #
like more classic consumer “pay-per-click” performance pageviews by popular category or
marketing reach / impressions / frequency by
● Conversations and conviction are important! target demographic
● Can incorporate outbound email tests after developing ● Refined numerical assumptions
more qualified lead list around CAC (CTR %, Cost per
qualified lead, etc.) build more robust
unit economics
DATA COLLECTION
COLLECT MORE DATA, AT SCALE

Example outputs:
Identify what can be done manually… your time is

● UX / Onboarding flows are
valuable!
straightforward to collect and help to
● Leverage freelancers for data collection understand current flow of user
● Automate / script data collection when possible experience and points of funnel
(open APIs, datasets) conversion
● For cold outbound, Identify which platforms would ● Product deep dive on specs to evaluate
be most relevant to your idea to scrape larger tech enablement
volume of leadgen; for example, ZoomInfo or ● Collect pricing info and codify to better
RocketReach compare what competitors are prioritizing
in product build
WHAT RESOURCES TO LEVERAGE FOR EACH HYPOTHESIS
Secondary Primary
Category Key criteria
Research Research
Big enough to build $1b+ company (conservatively assuming with a

Market Size
small share of pie and a small take rate)
✓ ✓
No major direct competitors (if winner-take-all situation) ✓
Competition
Long-term moat to protect against new entrants (network effects) ✓
High fragmentation in market ✓
Fragmentation
Scalable way to aggregate & vet / manage quality ✓
Clear value prop for supply ✓
Value Prop
Clear value prop for demand ✓
Business model with clear monetization opportunities ✓
Unit Economics High AOV and/or high frequency of purchases ✓ ✓
Realistic path to achieve net LTV/CAC @12mo ≥2x ✓ ✓
Repeatable & scalable acquisition engine ✓
Acquisition
If B2B <1 yr sales cycles ✓
A FEW EXAMPLES
A FEW EXAMPLES
2 EXAMPLES OF HYPOTHESES VALIDATION
Criteria High-level
Is this a big enough bottoms-up opportunity
Market Size
to build $1b+ company?
Do you have competitive differentiation &
Competition
sustainable moat?
Is supply highly fragmented, and can you
Fragmentation
aggregate & vet / manage quality?
Does this address clear value props that B2B beauty manufacturing
Value Prop
meet acute customer pain points?
Can you get to unit economics net
Unit Economics
LTV/CAC @12mo ≥2x?
Can you build scalable acquisition
Acquisition
channels, <1 yr sales cycles if B2B?
A FEW EXAMPLES
2 EXAMPLES OF HYPOTHESES VALIDATION
Criteria High-level
Market Size
Competition
sustainable moat?
Fragmentation
Value Prop
Unit Economics
Acquisition
Example of what
we’ve invalidated
Example of what
we’ve validated
HYPOTHESIS & VALIDATION OVERVIEW FOR BEAUTY MANUFACTURING IDEA
Criteria High-level
Market Size
Competition
sustainable moat?
Fragmentation
Value Prop
Unit Economics
Acquisition
Example of what
we’ve invalidated
Example of what
we’ve validated
B2B Beauty
Value Prop
HYPOTHESIS & VALIDATION OVERVIEW FOR BEAUTY MANUFACTURING IDEA Manufacturing
Identify key hypotheses Test hypotheses
Our initial hypothesis What we tested Where we landed
“Beauty brands and vendors would Primary Research “Most beauty manufacturers would not
be open to implementing an be willing to implement and improved
● Conversations
improved workflow solution” workflow”
● Performance marketing
● They are unhappy with the current tests ● Incredibly old school; hard to get them
process ● Outbound email tests onto a SaaS platform
● Lack of transparency actually works in
their favor (e.g., it's almost guaranteed
they're going to miss their deadlines)
B2B Beauty
Value Prop
EXAMPLE: OUTBOUND EMAIL TESTS Manufacturing
Test emails to potential customers + run tests on Interviews (not pictured)

response rate and willingness to pay
HYPOTHESIS & VALIDATION OVERVIEW FOR SEAFAIR
Criteria High-level
Market Size
Competition
sustainable moat?
Fragmentation
Does this address clear value props that Beauty manufacturing co.
Value Prop
Unit Economics
Acquisition
Example of what
we’ve invalidated
Example of what
we’ve validated
Unit Economics
HYPOTHESIS & VALIDATION OVERVIEW FOR SEAFAIR
Identify key hypotheses Test hypotheses
Our initial hypothesis What we tested Where we landed
“A marketplace connecting shipping Secondary Research “Crewing 1K vessels (2% of market)

companies with vetted seafarers will leads to $500M GMV and $25M CM3
● Open APIs and public data
have attractive unit economics” annually”
sets
● High AOV and retention are key Primary Research ● Profitable at 1st order and strong
drivers expected retention
● Conversations
● Clear path to monetization ● At scale they have the potential to
● SEM analyses
become extremely lucrative
● Performance marketing
tests
Unit Economics
EXAMPLE: UNIT ECONOMICS FOR SEAFAIR
Tested Google ads + landing page testing Captured Cost per Application for high-intent
supply to refine Supply CAC assumptions
Google Ads in targeted geo (Philippines) Ads led to landing page with
application form tracked in Airtable
Unit Economics
EXAMPLE: UNIT ECONOMICS FOR SEAFAIR
Over time, our unit economics assumptions evolved as

we refined key inputs
started with v1 early unit

economics for Y1-2
after a few weeks, v2 projected

unit economics at scale
Appendix
WHAT IS A WELL-DEFINED HYPOTHESIS?
Keep in mind HYPOTHESES SHOULD MEET KEY CRITERIA:
Hypotheses are predictions

Criteria High-level
• “I believe [X customer, player, or market] will [do Y action]
because of [Z reason]” Is this a big enough bottoms-up
Market Size
• Draw a line in the sand based on initial research… it’s OK opportunity to build $1b+ company?
if it’s just a guess as your starting point! Competition
Do you have competitive
differentiation & sustainable moat?
• For better prioritization as you develop testing plan:
Is supply highly fragmented, and can
○ “How confident am I based on existing research?” Fragmentation
you aggregate & vet / manage quality?
○ “If untrue, does this break the business idea?” Does this address clear value props
Value Prop
Must be testable that meet acute customer pain points?
• You will continuously refine hypotheses over time to Can you get to unit economics net
Unit Economics
include important findings and numbers LTV/CAC @12mo ≥2x?
• if possible, include baseline goals that seem achievable Acquisition
(to quantify value creation from your idea)
Validation mindset
• It should be well-structured and reasonable
• Approach with self-awareness, diligence, and data
AFTER YOU DEFINE HYPOTHESES
PRIORITIZE YOUR LIST FORMULATE A PLAN
Stack rank what to investigate first Set benchmarks

• Begin with what hypotheses would • For each hypothesis, understand baseline and what
absolutely break your model “proof point” you would need to get comfortable
• Start with low confidence but assumption • Quantify what proving vs. disproving means, if possible
could make or break the business
Design tests
• 2x2 matrix might be helpful: Level of
• Define optimal validation channels (can you get
confidence vs. Level of importance
comfortable with assumptions and/or is it possible to
collect enough volume through different methods)
VALIDATION EXAMPLE
What is the idea?
IDEA: JOB PLATFORM FOR THE VETERINARY INDUSTRY
Trends Our initial idea
The US pet industry is growing rapidly Create a labor platform for the
at over $95B veterinary industry
● Veterinary spend is one of the biggest ● Connect veterinarians with work
shares of the pet industry opportunities at vet clinics and
hospitals
● Largest part of a vet’s cost structure
is labor (~40-50%)
IDEA: JOB PLATFORM FOR THE VETERINARY INDUSTRY
What needs to be true?
KEY CRITERIA
Criteria
? Clear value prop on both sides
? Business model with clear monetization opportunities
? Unit economics that work
? Either high AOV or high frequency of purchases
? High fragmentation in market (important for a marketplace biz)
? No major direct competitors (if a winner take all situation)
? <1 year sales cycle
? Competitive moat / differentiation against new entrants
? TAM is big enough (conservatively assuming with a small share of pie and a small take rate)
KEY CRITERIA
Criteria
? Clear value prop on both sides
CLEAR VALUE PROP ON BOTH SIDES
Pain point hunting initially: Facebook, job boards and informal conversations
Pain point hunting at scale

Pain point hunting at scale

KEY CRITERIA
Criteria
✓ Clear value prop on both sides
BUSINESS MODEL W CLEAR MONETIZATION OPPORTUNITIES
BUSINESS MODEL W CLEAR MONETIZATION OPPORTUNITIES
UNIT ECONOMICS THAT WORK
“Back of the envelope” unit economics
AOV (GMV per order): $1,000 for 1 day shift

Take rate: 20%
= Revenue per order: $200
(-) COGS (e.g., Shipping, packaging): $0
= CM1 (Gross Margin): $200
Non COGS variable costs:

(-) Landing costs: logistics, packaging, handling, fulfillment
(-) Customer service (e.g. call center agents or reps)
(-) Payment: (e.g. credit card fees)
= CM2
Paid CAC:
(-) Offline marketing costs
(-) discounts , coupons
= CM3
(-) Fixed costs (e.g. salaries, rent, other overhead)
= EBITDA
KEY CRITERIA
Criteria
✓ Business model with clear monetization opportunities
✓ Unit economics that work
✓ Either high AOV or high frequency of purchases
HIGH FRAGMENTATION
Source: IBIS data

KEY CRITERIA
Criteria
✓ High fragmentation in market (important for a marketplace biz)
NO MAJOR DIRECT COMPETITORS
Very poor existing recruiting options
Where to find a vet when hiring

•American Veterinary Medical Association (AVMA) or local VMA: $300-$500 for a posting
•Word of mouth / social media: very inefficient, not scalable
•Staffing agency: very fragmented, high fees, not scalable
•Public job boards (Indeed, Ziprecruiter, iHireVeterinary): Extremely poor-quality candidates,
most applicants haven’t even gone to vet school
KEY CRITERIA
Criteria
✓ No major direct competitors (if a winner take all situation)
<1 YR SALES CYCLE
<1 YR SALES CYCLE
Sales test for short-term vet shifts
● Validating supply-side sales cycle via calls
○ Tested 4 different value props
○ #1 value prop: Job discovery
■ Cold outbound to 150 contacts, with a
full sales campaign
● (LinkedIn, phone call, 5 emails)
■ Had 7 successes (waitlist & shift
availability), or 5% conversion rate

KEY CRITERIA
Criteria
✓ <1 year sales cycle
COMPETITIVE MOAT / DIFFERENTIATION AGAINST NEW ENTRANTS
Our hypothesis
Start by building liquidity on a city-by-city basis, add tool / features

● There is currently no aggregated pool of supply or demand for the veterinary
industry
● Long-term we could provide tools such a tax forms and insurance to prevent
disintermediation
● We could provide a superior customer experience than alternatives
KEY CRITERIA
Criteria
✓ Competitive moat / differentiation against new entrants
IS TAM BIG ENOUGH?
IS TAM BIG ENOUGH?
KEY CRITERIA
Criteria
✓ Competitive moat / differentiation against new entrants
X TAM is big enough (conservatively assuming with a small share of pie and a small take rate)
CONCLUSION
Small TAM can be a good thing if:
You can increase the “size of the pie”

• Do you reasonably believe that your solution will enable you to grow TAM?
• Is there a way to unlock “dormant” demand and/or supply?
• Example:
You can eventually go into a conjoined category

• Once you’ve won your category can you leverage your supply or demand
to expand into another area / industry?
• Example:

Validating Your Startup Idea

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Validating Your Startup Idea

Uploaded by

Copyright:

Available Formats

VALIDATING YOUR STARTUP IDEA

Ideate Get inspired and come up with ideas

Litmus test See which ideas are worth pursuing further

Identify potential business models, create

hypotheses to test validity of models

Try to prove or disprove each statement,

It’s critical to:

Ideate Get inspired and come up with ideas

Litmus test See which ideas are worth pursuing further

Identify potential business models, create

hypotheses to test validity of models

Try to prove or disprove each statement,

It’s critical to:

HYPOTHESES SHOULD MEET KEY CRITERIA:

Category Key question Key criteria

Do you have competitive No major direct competitors (if winner-take-all situation)

Framework Understand critical levers

AOV (GMV per order) • In your model, understand the most

Ideate Get inspired and come up with ideas

Litmus test See which ideas are worth pursuing further

Identify potential business models, create

hypotheses to test validity of models

Try to prove or disprove each statement,

It’s critical to:

Secondary Research Primary Research

● Conducted by others ● Self-conducted

● S-1s, 10-Ks, etc ● Conversations, interviews

HIGH-LEVEL MORE GRANULAR

• Customer Discovery: conversations, interviews, and surveys

INTERVIEWS & SURVEYS

PERFORMANCE MARKETING & OUTBOUND TESTS

Analysis, then paid tests

COLLECT MORE DATA, AT SCALE

Identify what can be done manually… your time is

Big enough to build $1b+ company (conservatively assuming with a

2 EXAMPLES OF HYPOTHESES VALIDATION

2 EXAMPLES OF HYPOTHESES VALIDATION

Identify key hypotheses Test hypotheses

Our initial hypothesis What we tested Where we landed

Test emails to potential customers + run tests on Interviews (not pictured)

Identify key hypotheses Test hypotheses

Our initial hypothesis What we tested Where we landed

“A marketplace connecting shipping Secondary Research “Crewing 1K vessels (2% of market)

Over time, our unit economics assumptions evolved as

started with v1 early unit

after a few weeks, v2 projected

Keep in mind HYPOTHESES SHOULD MEET KEY CRITERIA:

Hypotheses are predictions

PRIORITIZE YOUR LIST FORMULATE A PLAN

Stack rank what to investigate first Set benchmarks

Trends Our initial idea

? Clear value prop on both sides

? Business model with clear monetization opportunities

? Unit economics that work

? Either high AOV or high frequency of purchases

? High fragmentation in market (important for a marketplace biz)

? No major direct competitors (if a winner take all situation)

? <1 year sales cycle

? Competitive moat / differentiation against new entrants

? Clear value prop on both sides

? Business model with clear monetization opportunities

? Unit economics that work

? Either high AOV or high frequency of purchases