Experimentation, Validation and Lean Start-up

Introduction and Context

Entrepreneurship is often described as the creation of new organizations, but entrepreneurial work begins earlier: with decisions made under uncertainty. Before a market, business model, or technology has been proven, founders must decide what to believe, what to test, and how much to invest. The central managerial challenge is not simply generating ideas; it is learning which assumptions deserve further commitment.

The lean start-up approach provides a disciplined method for addressing this challenge. It combines hypothesis-driven entrepreneurship, rapid experimentation, minimum viable products (MVPs), customer development, and innovation accounting. Its central proposition is that a new venture should treat its business model as a set of provisional hypotheses rather than as an established plan. The venture then conducts carefully designed tests, gathers evidence, and uses that evidence to persevere, pivot, or terminate a concept.

Lean start-up methods do not mean building carelessly, avoiding strategy, or accepting every customer statement as proof. They require rigorous problem definition, explicit assumptions, valid measures, ethical research, and managerial judgment. At graduate level, the approach should therefore be understood as a decision system for learning under uncertainty—not as a collection of shortcuts.

Learning Objectives

By the end of this chapter, you should be able to:

1. Distinguish facts, assumptions, hypotheses, and risks in an entrepreneurial opportunity.
2. Convert business-model assumptions into testable hypotheses.
3. Design appropriate experiments and minimum viable products.
4. Select metrics that measure learning rather than superficial activity.
5. Interpret evidence using innovation accounting and cohort analysis.
6. Decide whether to persevere, pivot, pause, or terminate a concept.
7. Evaluate the ethical, methodological, and strategic limitations of lean experimentation.

1. Entrepreneurship as Hypothesis-Driven Learning

A traditional business plan typically presents a coherent forecast of customers, revenue, costs, and growth. Such a plan can be useful for communicating an intended direction, but early-stage forecasts often rest on untested assumptions. A lean approach reverses the sequence: instead of treating the plan as a prediction to be executed, the entrepreneur treats it as a model to be tested.

A business-model assumption becomes useful only when it is expressed as a proposition that could be supported or contradicted by evidence. For example:

• Vague assumption: “Small retailers need better inventory software.”
• Testable hypothesis: “At least 30% of independent retailers contacted in the target segment will identify stock-outs as one of their three most costly operational problems.”
• Stronger behavioral hypothesis: “At least 20 of 50 qualified retailers will agree to a paid four-week pilot of a stock-out alert service.”

The second and third formulations specify a population, behavior, and threshold. They make disagreement possible, which is essential for learning.

Facts, assumptions, hypotheses, and risks

These terms should not be treated as synonyms.

Entrepreneurs should prioritize assumptions that are both important and uncertain. A minor assumption may not deserve an experiment, while a highly uncertain assumption can justify substantial investigation if its failure would destroy the business model.

The business model as a system of hypotheses

A business model contains interdependent assumptions. Common categories include:

• Customer problem: Is the problem significant, frequent, and costly enough to motivate action?
• Customer segment: Who experiences the problem and who has authority to decide or pay?
• Value proposition: Does the proposed solution create meaningful customer value?
• Channel: Can the venture reach customers effectively?
• Revenue model: Will customers pay, how much, and when?
• Cost structure: Can the venture deliver the value proposition at sustainable cost?
• Key resources and activities: Can the venture obtain the capabilities required for delivery?
• Partners: Are suppliers, platforms, regulators, or distributors necessary?
• Competitive response: Can the venture remain differentiated when incumbents respond?

The Business Model Canvas is often used to visualize these assumptions. However, the canvas is not evidence. It is a map of what must be learned.

2. From Assumptions to Testable Hypotheses

A well-formed hypothesis normally specifies five elements:

1. Unit of analysis: the customer, transaction, account, or user being observed.
2. Target population: the relevant customer segment.
3. Behavior or outcome: what the subject will do, not merely what the subject will say.
4. Context and time period: when and under what circumstances behavior is measured.
5. Threshold: the result that will count as sufficient evidence.

A useful structure is:

Among [target population], at least [threshold] will [observable behavior] within [specified context and time period], because [proposed causal mechanism].

For example:

Among independent cafés with at least two locations, at least 25% of qualified decision-makers will authorize a two-week paid trial of a demand-forecasting service within 30 days, because stock waste is a sufficiently costly problem.

The phrase “will authorize a paid trial” is stronger than “will express interest.” Expressions of interest are valuable discovery evidence but may not predict adoption or payment.

Prioritizing assumptions

A practical prioritization matrix considers:

• Impact: How damaging would failure be?
• Uncertainty: How little credible evidence currently exists?
• Testability: How quickly and ethically can the assumption be investigated?

A high-impact, high-uncertainty assumption is a critical risk. A venture should normally test it before investing heavily in product development, hiring, or infrastructure.

A more formal decision perspective treats an experiment as valuable when it can change a consequential decision. The value of information depends not only on the amount of data collected but also on whether the information could alter resource allocation. Ten additional interviews may have low value if the venture has already established the problem clearly but has not tested willingness to pay.

3. Experiment Design for New Ventures

An experiment is a structured investigation in which an entrepreneur exposes a defined subject or market situation to a stimulus and observes a specified outcome. The experiment should answer a decision question, not merely generate activity.

Components of a strong experiment

A practical experiment brief should contain:

• Decision: What decision will the result inform?
• Hypothesis: What proposition is being tested?
• Method: What will be shown, offered, or changed?
• Sample: Who will participate and how will they be selected?
• Measure: What observable outcome will be recorded?
• Threshold: What result supports or challenges the hypothesis?
• Duration: When will the test start and stop?
• Controls: What alternative explanations must be considered?
• Ethics: What consent, privacy, safety, or fairness requirements apply?

For example, if a venture tests a landing page, the central outcome may be qualified sign-ups rather than page views. The page should represent a realistic value proposition, and the traffic source should resemble the intended acquisition channel. Otherwise, the experiment may test an artificial audience rather than the market.

Exploratory and confirmatory experiments

Early-stage entrepreneurship often requires exploratory research. Interviews, observation, diary studies, and open-ended prototypes help reveal customer language, workarounds, and unrecognized needs. These methods are particularly useful when the problem is poorly understood.

Confirmatory experiments test a more specific proposition. Examples include a pricing test, a pre-order, a concierge pilot, or an A/B comparison. Confirmatory tests require greater discipline because the entrepreneur has specified an expected outcome in advance.

The distinction is important. An interview can reveal that a problem exists, but it cannot by itself establish market size, conversion rates, or profitability. Conversely, a low conversion rate may indicate a poor message, an unsuitable channel, weak trust, or an unimportant problem. Evidence must be interpreted in context.

Causal validity and common threats

Entrepreneurs should distinguish correlation from causation. If sales rise after a new advertisement is launched, the increase may result from seasonality, a competitor’s failure, a price change, or existing word-of-mouth rather than the advertisement.

Common threats include:

• Selection bias: participants differ systematically from the target market.
• Response bias: participants provide answers they believe are socially desirable.
• Survivorship bias: analysis includes successful users but excludes people who abandoned the product.
• Novelty effects: early enthusiasm fades after initial exposure.
• Confounding: multiple changes occur simultaneously.
• Small samples: random variation is mistaken for a stable pattern.
• Instrumentation changes: the measurement method changes during the test.

Randomization, consistent measurement, comparison groups, repeated observation, and transparent documentation can improve confidence. Early-stage experiments rarely achieve laboratory-level certainty, but they can still reduce uncertainty sufficiently for a decision.

4. Minimum Viable Products

An MVP is the simplest product, service, or representation that enables a venture to test a critical assumption with real or realistically committed users. “Minimum” refers to the smallest learning investment, not merely the smallest feature set. “Viable” means sufficiently credible and functional for the intended test; it does not mean that the entire business is already viable.

An MVP may take many forms:

An MVP is not permission to deceive customers, expose them to unsafe conditions, or disregard legal obligations. If a manual process is presented as automated, the venture must avoid materially misleading claims and should disclose relevant limitations when they affect customer decisions.

MVP scope and learning objectives

The appropriate MVP depends on the riskiest assumption. If the risk is usability, an interactive prototype may be sufficient. If the risk is willingness to pay, a polished prototype with no payment request is inadequate. If the risk is operational reliability, a landing page is insufficient.

This yields a basic design rule:

Build only what is necessary to produce credible evidence about the most consequential uncertainty.

The rule does not mean that product quality is irrelevant. Poor reliability, confusing design, or an implausible offer can create false negative results. The MVP must be credible enough that failure reflects the hypothesis rather than obvious defects unrelated to the value proposition.

5. The Build–Measure–Learn Loop

The lean start-up cycle is commonly expressed as Build–Measure–Learn. The sequence is iterative rather than strictly linear:

1. Build: create an experiment or product increment.
2. Measure: observe behavior using defined metrics.
3. Learn: interpret the evidence and update assumptions.
4. Decide: persevere, pivot, pause, or terminate.

The loop becomes ineffective when the venture builds continuously without testing, measures only activity, or “learns” without changing a decision. Learning should be documented as a change in confidence, a revised hypothesis, or a decision about resource allocation.

Innovation accounting

Innovation accounting adapts measurement to ventures whose conventional financial statements do not yet reveal meaningful performance. It generally involves three layers:

1. Establish the baseline: measure the current behavior of a defined cohort.
2. Tune the engine: improve the drivers of acquisition, activation, retention, revenue, or referral.
3. Decide whether to pivot or persevere: compare progress with the venture’s thresholds and strategic alternatives.

A useful funnel might contain:

• qualified visitors,
• activated users,
• retained users,
• paying customers,
• repeat transactions, and
• referred customers.

Vanity metrics—such as total downloads, registered users, or social-media impressions—can appear impressive while concealing poor retention or non-paying usage. Actionable metrics are tied to a decision and defined for a specific population and period.

Core quantitative metrics

For a specified period:

The numerator and denominator must be defined consistently. If 80 of 400 qualified visitors request a trial, conversion is 20%.

Retention for a cohort can be represented as:

A cohort is a group sharing a meaningful starting event, such as signup in the same week. Cohort retention is generally more informative than the total number of active users because it distinguishes growth from continued use.

If customer acquisition cost is estimated over a period:

CAC is meaningful only when the cost period, attribution rule, and customer count are aligned. It can be misleading when founder labor, discounts, channel fees, or onboarding costs are omitted.

For a subscription business, a simplified contribution measure is:

This is not the same as accounting profit. Fixed costs, research expenditure, taxes, financing costs, and opportunity costs may remain substantial.

A simplified payback period, assuming stable monthly contribution, is:

This expression applies only when CAC and contribution are measured on a comparable basis and monthly contribution is positive and reasonably stable. If monthly contribution is zero or negative, payback is not achieved through the assumed model.

#### Worked example: interpreting a cohort funnel

An illustrative venture acquires 500 qualified visitors in March. Of these, 125 activate the product, 50 remain active after four weeks, and 20 pay for a plan. The metrics are:

• Activation rate = 125 ÷ 500 × 100% = 25%.
• Four-week retention among activated users = 50 ÷ 125 × 100% = 40%.
• Paid conversion from qualified visitors = 20 ÷ 500 × 100% = 4%.
• Paid conversion from activated users = 20 ÷ 125 × 100% = 16%.

The results suggest that the venture has an activation problem, a retention challenge, or both. Increasing traffic alone may not solve the underlying problem. The next experiment should investigate why activated users fail to experience continuing value, while also checking whether the paying group differs from non-paying users.

Metrics should not be interpreted as universal benchmarks. A 4% conversion rate may be excellent in one category and poor in another. The correct comparison is with the venture’s prior cohorts, credible competitors, unit economics, strategic objectives, and predefined thresholds.

6. Validated Learning and the Pivot Decision

Validated learning is evidence-based knowledge about which elements of the business model are working, for whom, and under what conditions. It is not simply the accumulation of data. Learning is validated when an experiment tests a meaningful assumption with a credible measure and produces a conclusion that guides action.

Persevere, pivot, pause, or terminate

A persevere decision is appropriate when evidence supports the core hypothesis and the remaining problems are addressable through execution or optimization.

A pivot is a substantial, intentional change to one or more business-model elements while preserving a learning objective or broader strategic direction. Examples include:

• customer-segment pivot,
• customer-need pivot,
• channel pivot,
• revenue-model pivot,
• technology pivot,
• platform pivot,
• value-capture pivot, and
• zoom-in or zoom-out pivot.

A pause may be appropriate when evidence is inconclusive, resources are temporarily constrained, or external conditions make the test unreliable. A terminate decision is appropriate when a critical hypothesis is repeatedly contradicted, the economics are structurally unattractive, or ethical, legal, or operational barriers cannot be responsibly overcome.

The decision should be based on predefined criteria where possible. For example:

A pivot is not automatically a success. It is a new hypothesis that must be tested. Repeatedly changing direction without preserving learning creates strategic drift and makes accountability difficult.

7. Case Study: FarmLink Produce Exchange

The following is a fictional illustrative case designed for educational analysis.

Background

FarmLink is a proposed digital marketplace connecting small urban restaurants with regional vegetable growers. The founding team believes restaurants want fresher produce, while growers need better demand visibility. The initial business model assumes that restaurants will pay a 10% transaction fee for access to a reliable supply of locally grown produce.

Problem and initial assumptions

The founders identify three critical assumptions:

1. Restaurant chefs experience frequent difficulty sourcing consistent local produce.
2. Chefs will place orders through a digital marketplace rather than through existing wholesalers.
3. Transaction margins will cover delivery, quality assurance, customer support, and acquisition costs.

The team initially proposes building a full marketplace with inventory synchronization, payment processing, route optimization, and supplier analytics. A risk review shows that the most dangerous unknown is not technical feasibility but whether restaurants will change purchasing behavior.

Experiment design

The team conducts 30 interviews with owners and purchasing managers from independent restaurants. Interviews reveal that freshness matters, but reliability and delivery timing matter more. The team then creates a concierge MVP: for four weeks, it manually collects weekly orders, matches them with five farms, and coordinates delivery using a shared ordering form and telephone confirmation.

The pre-specified hypothesis is:

At least 8 of 20 qualified restaurants will place two or more paid orders during the four-week pilot if the service provides reliable delivery windows and item availability.

The team defines “qualified” as an independent restaurant purchasing at least 40 kilograms of vegetables per week and located within the delivery area. It records orders, gross merchandise value, delivery cost, substitutions, complaints, and repeat behavior.

Results

Twenty restaurants join the trial. Twelve place at least one order, but only six place two or more orders. The strongest dissatisfaction concerns substitutions and late delivery. The restaurants value local sourcing but do not regard the marketplace as reliable enough for core menu planning. Delivery cost per order is also higher than expected because orders are geographically dispersed.

The experiment does not validate the original marketplace hypothesis. However, it provides several forms of learning:

• The problem is real but includes a stronger reliability requirement than expected.
• The initial customer segment may be too broad; restaurants with flexible seasonal menus are more tolerant of substitutions.
• A transaction-fee model may be insufficient if logistics remain expensive.
• Manual coordination reveals operational requirements that software alone would not solve.

Pivot decision

The team rejects a full marketplace build and tests a narrower model: scheduled weekly produce boxes for restaurants with seasonal menus, supplied by a smaller geographic cluster of farms. The value proposition changes from “access to local supply” to “predictable weekly local produce with transparent substitutions.” The team also explores a subscription fee paid by restaurants and a shared delivery route.

Outcome and learning points

In the second pilot, repeat ordering improves, but the team discovers that restaurants want invoices and quality guarantees more than consumer-style marketplace features. FarmLink eventually focuses on institutional buyers—small schools and workplace cafeterias—that can plan menus several weeks ahead. The case illustrates that:

1. Customer enthusiasm is weaker evidence than repeated paid behavior.
2. An MVP should test the riskiest assumption, not showcase every planned feature.
3. Operational constraints can invalidate an apparently attractive digital model.
4. A pivot should preserve learning while changing the business model deliberately.
5. Evidence can support a more focused opportunity rather than simply confirming or rejecting the original idea.

8. Ethical, Legal, and Governance Considerations

Experimentation involves real people, data, and consequences. Entrepreneurial speed does not remove responsibilities concerning privacy, safety, fairness, and truthful communication.

Informed participation and truthful claims

Participants should understand what they are being asked to do, what data will be collected, and whether they will receive compensation or access to a prototype. A fake-door test—where a user is invited to click an unavailable feature—may measure interest, but it should not collect payment or create a false impression that a service is immediately available. If the test affects a customer’s decision materially, disclosure is ethically important.

Data protection

Experiments should collect only data necessary for the stated purpose, restrict access, define retention periods, and protect sensitive information. Personal data should not be repurposed casually. Depending on jurisdiction and sector, privacy laws may impose requirements concerning consent, lawful processing, data minimization, security, deletion, and automated decision-making.

Fairness and algorithmic risk

AI-enabled MVPs can produce discriminatory outcomes if training data reflect historical bias or if performance differs across groups. A promising average conversion rate does not justify deploying a system that systematically denies services to protected populations. Teams should test subgroup performance, document limitations, provide human review where appropriate, and avoid claiming reliability beyond the evidence.

Safety and regulated domains

Health, finance, education, mobility, food, and employment experiments may require professional oversight or regulatory approval. An MVP cannot be used to evade product-safety rules, financial regulations, clinical requirements, or professional standards. In high-risk contexts, the cost of a false negative or false positive may be much greater than the cost of a failed business experiment.

Governance of entrepreneurial evidence

Founders should maintain an experiment log recording the original hypothesis, date, sample, method, results, deviations, interpretation, and decision. This reduces hindsight bias and prevents selective reporting. Investors and employees should be able to distinguish observed facts from management interpretation and aspiration.

9. Limitations and Critical Evaluation

Lean start-up methods are influential but not universally sufficient.

First, experiments can overemphasize measurable short-term behavior. Radical innovations may require education, complementary infrastructure, institutional change, or long adoption periods. Early low usage does not always imply low potential.

Second, customer experimentation can privilege incremental problems that are easy to test. Social ventures and sustainability ventures may create benefits that are distributed across communities and future generations rather than captured by an immediate paying customer.

Third, local evidence may not generalize. A pilot with technology-oriented early adopters may not predict behavior among mainstream customers. External validity must be considered when moving from a niche cohort to a larger market.

Fourth, repeated pivots can create path dependence or strategic confusion. The team may optimize a metric while losing sight of its purpose. Metrics are representations of value, not value itself.

Fifth, the framework can understate power relationships. Customers, employees, suppliers, regulators, and communities may have conflicting interests. A hypothesis can be commercially attractive but socially harmful.

Finally, experimentation does not eliminate uncertainty. It changes the distribution of confidence about possible outcomes. Entrepreneurial judgment remains necessary because decisions involve opportunity costs, irreversible commitments, timing, ethics, and strategic positioning.

10. Best Practices for Graduate-Level Application

A disciplined experimentation program should:

1. Begin with the decision, not the preferred method.
2. Identify the most consequential uncertain assumption.
3. State an observable hypothesis and a threshold before collecting evidence.
4. Prefer behavior, payment, retention, and operational performance over opinions alone.
5. Match the MVP to the specific risk being tested.
6. Use cohorts and consistent definitions for metrics.
7. Separate discovery evidence from validation evidence.
8. Record negative results and unexpected findings.
9. Test operational and economic assumptions, not only customer appeal.
10. Protect participants and communicate limitations honestly.
11. Set a time or resource boundary for each experiment.
12. Make pivot criteria explicit and review them with the team.
13. Consider whether evidence transfers to the intended market.
14. Use AI and analytics to accelerate analysis, not to replace customer judgment or ethical oversight.
15. Stop experiments that cannot produce a consequential decision.

Chapter Summary

Experimentation and validation provide a disciplined response to entrepreneurial uncertainty. A venture should treat its business model as a set of assumptions and convert the most important assumptions into testable hypotheses. Strong hypotheses specify a population, behavior, context, time period, and threshold.

Experiments should be designed around decisions, use credible measures, and account for selection bias, response bias, confounding, and limited generalizability. An MVP is the smallest credible artifact or service that tests a critical assumption; it is not simply a product with few features. The Build–Measure–Learn cycle is valuable only when measurement produces validated learning and changes a decision.

Innovation accounting emphasizes actionable metrics such as activation, cohort retention, conversion, customer acquisition cost, and contribution margin. These metrics require precise definitions and should not be confused with vanity measures. Evidence may justify perseverance, a deliberate pivot, a pause for further investigation, or termination of a concept.

Lean experimentation is not a substitute for ethics, strategy, regulation, or managerial judgment. Privacy, truthful communication, fairness, safety, and governance remain essential. Its greatest contribution is methodological: it helps entrepreneurs invest progressively, learn from real behavior, and allocate scarce resources toward opportunities supported by credible evidence.

Key Terms

Assumption: A belief used in constructing or operating a business model.

Business Model Canvas: A visual framework for representing key components of a business model.

Cohort: A group of users or customers sharing a meaningful starting event or characteristic.

Concierge MVP: A manually delivered service used to test customer value before automation.

Conversion rate: The percentage of eligible opportunities that result in a specified desired action.

Customer development: A process of discovering and validating customer problems, solutions, and business models.

Experiment: A structured investigation designed to produce evidence about a hypothesis.

Hypothesis: A specific, testable proposition about expected behavior or outcomes.

Innovation accounting: A measurement approach for tracking progress in uncertain, early-stage ventures.

Lean start-up: An entrepreneurial approach centered on rapid experimentation, validated learning, and iterative business-model development.

Minimum viable product (MVP): The smallest credible product or service capable of testing a critical assumption.

Pivot: A substantial, intentional change in business-model direction based on learning.

Retention rate: The proportion of an original cohort that remains active at a specified time.

Selection bias: Systematic distortion caused by how participants or observations are selected.

Threshold: A pre-specified result used to interpret whether a hypothesis is supported.

Validated learning: Evidence-based knowledge that reduces uncertainty and informs a decision.

Vanity metric: A measure that appears positive but provides little guidance for action.

Wizard-of-Oz MVP: A prototype in which manual operations produce an experience that appears technologically automated to the user.

Review Questions

Short Answer Questions

1. What distinguishes an assumption from a testable hypothesis?
2. Why is repeated paid behavior generally stronger evidence than stated interest?
3. Define an MVP and explain why it is not necessarily a low-quality product.
4. What is the purpose of cohort analysis?
5. Name two ethical responsibilities involved in entrepreneurial experimentation.

Long Answer Questions

1. Explain how a business-model assumption can be converted into a hypothesis, using an original example.
2. Discuss the Build–Measure–Learn cycle and identify conditions under which it may fail.
3. Compare exploratory and confirmatory experiments in early-stage venture development.
4. Explain innovation accounting and distinguish actionable metrics from vanity metrics.
5. Critically evaluate the strengths and limitations of the lean start-up approach for a social or regulated venture.

Analytical / Application Questions

1. A start-up obtains many website visits but very low four-week retention. Design an experiment to determine whether the problem is poor activation, weak ongoing value, or an unsuitable customer segment.
2. A venture’s CAC is $120 and its monthly contribution margin per customer is $30. Calculate the simplified payback period and identify two assumptions that must hold for the calculation to be meaningful.
3. Using the FarmLink case, propose a new hypothesis, MVP, metric, and pivot criterion for testing the institutional-buyer segment.

Multiple Choice Questions with Answers

Question 1

Which statement is the strongest example of a testable entrepreneurial hypothesis?

A. Customers will love our product.

B. The market for this product is large.

C. At least 25% of qualified restaurants will place a paid repeat order within four weeks.

D. Our technology is innovative.

Correct Answer: C. At least 25% of qualified restaurants will place a paid repeat order within four weeks.

Question 2

What is the primary purpose of a minimum viable product?

A. To provide every feature planned for the final product

B. To test a critical assumption with the smallest credible investment

C. To eliminate the need for customer research

D. To maximize short-term website traffic

Correct Answer: B. To test a critical assumption with the smallest credible investment

Question 3

Which metric is most directly concerned with continued use by an original group of users?

A. Retention rate

B. Total downloads

C. Advertising impressions

D. Number of press mentions

Correct Answer: A. Retention rate

Question 4

A venture has 80 desired actions from 400 eligible opportunities. What is the conversion rate?

A. 5%

B. 20%

C. 25%

D. 320%

Correct Answer: B. 20%

Question 5

Which situation most clearly represents a pivot?

A. Increasing the server capacity without changing the business model

B. Correcting a spelling error in the user interface

C. Changing from selling to individual users to selling to institutional buyers based on evidence

D. Repeating the same experiment without changing its design

Correct Answer: C. Changing from selling to individual users to selling to institutional buyers based on evidence

Question 6

Why can total registered users be a vanity metric?

A. It always measures profitability accurately.

B. It may increase without showing activation, retention, or payment.

C. It cannot be calculated for digital products.

D. It is only useful for government organizations.

Correct Answer: B. It may increase without showing activation, retention, or payment.

Question 7

A venture spends $12,000 to acquire 100 new customers. Using the stated CAC formula, what is CAC?

A. $12

B. $120

C. $1,200

D. $12,000

Correct Answer: B. $120

Question 8

Which is a common threat to the validity of an entrepreneurial experiment?

A. Predefining a decision threshold

B. Recording the sample and measurement period

C. Selection bias in recruiting participants

D. Using a clearly specified behavioral outcome

Correct Answer: C. Selection bias in recruiting participants

Question 9

If CAC is $120 and monthly contribution margin is $30, what is the simplified payback period?

A. 2 months

B. 3 months

C. 4 months

D. 6 months

Correct Answer: C. 4 months

Question 10

Which action best reflects validated learning?

A. Collecting more data without linking it to a decision

B. Changing a business-model assumption after a credible experiment contradicts it

C. Reporting only successful customer interviews

D. Building additional features before measuring current usage

Correct Answer: B. Changing a business-model assumption after a credible experiment contradicts it