Hypothesis testing and MVP

In this section, we understand how to move from a filled Lean Canvas to validating a venture idea using hypotheses testing and Minimum Viable Products (MVPs).

From Assumptions to Evidence

After filling out a Lean Canvas, it's crucial to understand that everything documented on it is currently an assumption or hypothesis. The entire venture-building journey is about transforming these assumptions into evidence through validation. This involves moving from opinions to verified facts about the market, customers, problem, solution, and value proposition. It's important to be prepared to change course if assumptions are disproven.

The Build-Measure-Learn Loop

The core of testing hypotheses is the iterative "Build-Measure-Learn" loop:

• Build: Create something that encapsulates your assumptions or ideas. This doesn't necessarily mean a fully functional product; it could be a pitch, a video, a survey, or a basic prototype. The goal is to build just enough to test a specific hypothesis.
• Measure: Gather data on how customers respond to what you've built. This could involve tracking engagement, interest, or direct feedback.
• Learn: Analyze the data to determine if your assumptions were validated.
  • If validated, you "persevere" and continue on the current path, then test the next set of assumptions.
  • If not validated, you "pivot," meaning you make a course correction, potentially by redefining the customer segment, problem, or solution based on what you've learned. The faster you iterate through this loop, the quicker you can validate or invalidate your venture's direction.

Types of Hypotheses

• Value Hypotheses (Early Stage): At the very beginning, the focus is on testing whether there is genuine value in what you're proposing. This includes assumptions about:
  • Is the problem real?
  • Do enough customers acknowledge and resonate with this problem?
  • Does the proposed product/service effectively solve the problem?
  • Do customers care about the unique value proposition?
  • Validating these leads to Problem-Solution Fit.
• Growth Hypotheses (Later Stage): Once value hypotheses are validated and you move towards Product-Market Fit and beyond, the focus shifts to hypotheses around:
  • Customer willingness to pay and price sensitivity.
  • Scalability of the business.
  • Cost optimization and margins.
  • Usability and user engagement.

Formulating Testable Hypotheses

Hypotheses should be testable and quantifiable to ensure clear measurement and confident validation.

• Avoid vague statements: Instead of "Our product is loved by customers," which is hard to measure definitively, quantify it.
• Quantify your hypotheses: For example, "At least 50% of customers admit that this is a real problem." This allows for clear evaluation. The percentage threshold can be estimated based on market analysis (e.g., Serviceable Obtainable Market).
• Measure engagement: For an app, instead of "customers use my product regularly," a better hypothesis is "customers use our app two times a week or spend 20 minutes every day on my app." Quantified metrics lead to better insights for monetization or further development.

Methods for Testing Hypotheses

Various methods can be employed to test hypotheses, evolving as the venture progresses:

• In-depth Customer Interviews: Crucial in the initial stages for qualitative insights into customer likes, dislikes, preferences, and the context of their problem. Entrepreneurs must actively engage with potential customers.
• Surveys: Can be distributed on various platforms to gauge customer responses.
• Focus Groups: Bringing groups of customers together for discussions and product demos to assess reactions.
• Digital Campaigns (Pre-launch): Using videos, landing pages, or pre-order campaigns to assess interest (e.g., asking for email sign-ups or letters of intent).
• A/B Testing (Later Stage): For established products/websites, testing different versions of features or designs (e.g., website screens) to measure user response and optimize.

The process of experimentation is continuous; while early-stage engagement is more qualitative, later stages often involve larger numbers and more quantitative testing.

Avoiding Pitfalls: False Positives and False Negatives

• False Positive: A positive response that is not actually true in reality. This often happens if the sample group is not representative (e.g., relying solely on friends and family). To prevent this, ensure your sample truly represents the target customer segment.
• False Negative: A negative response when, in fact, there is a genuine market or interest. This can occur if the messaging or product offering is unclear or not well-communicated to the customer. If a false negative is suspected, re-evaluate communication and conduct more tests. If it's a true negative, a pivot is necessary.

Minimum Viable Product (MVP)

An MVP is the smallest set of activities needed to rigorously disprove a hypothesis. It's the vehicle for testing assumptions in the Build-Measure-Learn loop.

• Minimal: It includes only the bare essentials needed to convey the core idea and test a specific assumption, avoiding excessive development.
• Viable: It must be a true reflection of the product/service, allowing customers to comprehend its essence.

Types of MVPs:

• No-Product MVP / Smoke Test MVP: This involves testing demand without building a product.
  • Examples: A video demonstrating how a product would work (like Dropbox's initial video), a pitch deck, or a simple landing page asking for interest. The goal is to gauge customer reaction and interest in the concept.
• "Sell Before You Build" MVP: Collecting payments or pre-orders for a product that hasn't been built yet (e.g., Kickstarter campaigns). This validates customer willingness to pay.
• Concierge MVP: Delivering the service manually to a few customers to gain deep understanding and validate assumptions, without relying on automated technology. The entrepreneur acts as a "concierge" (e.g., Food on the Table founder manually helping customers with meal planning and grocery shopping). This is a cheap way to test viability.
• Wizard of Oz MVP: The customer experiences a seemingly fully functioning service, but the backend operations are entirely manual. It creates the illusion of automation (e.g., Zappos initially fulfilling shoe orders by buying from retail stores, or BigBasket starting with manual order fulfillment).
• Single Feature Product MVP: A product with only the most essential features required to deliver on the unique value proposition. This is a step towards a more developed product.

The overarching philosophy is to test hypotheses rapidly and cheaply, building only after demand and value have been validated. The Lean Canvas and hypotheses testing work in tandem: the canvas documents assumptions, and the Build-Measure-Learn loop (using MVPs) tests these assumptions, leading to continuous refinement of the business plan. This iterative process drives the venture from Problem-Solution Fit to Product-Market Fit and eventually to scaling. The achievement of Product-Market Fit is often a good time to seek funding, as it aligns with investors' goals.