Building Your MVP

Build Less, Learn More

The instinct when starting a company is to build something impressive. Fight that instinct. The goal of an MVP is not to impress. It is to learn.

By the end of this lesson, you will know how to identify your core assumption, build the minimum product to test it, and extract maximum learning from real customer interactions.

Quick Recall: In the previous lesson, we mapped the competitive landscape, sized the market, and created our positioning statement. Now we know what to build. The question is: what is the absolute minimum version of that product that still tests our hypothesis?

The MVP Mindset

An MVP is not a crappy version of your full product. It is a focused experiment.

What an MVP is:

The minimum thing you can build to test your riskiest assumption
A learning tool, not a launch product
Something real customers can use or interact with
Built in days or weeks, not months

What an MVP is not:

Version 1.0 of your final product
A proof of concept with no real users
A prototype you only show to friends
Anything that takes more than 4-6 weeks to build

Identifying Your Core Assumption

Every startup has one assumption that determines everything else. Find it:

My startup idea: [DESCRIBE IT]

Help me identify my core assumption:
1. What must be true about customer behavior for this to work?
2. What must be true about willingness to pay?
3. What must be true about our ability to deliver?
4. Which of these, if wrong, makes everything else irrelevant?
5. How could I test this assumption with the least effort?

Examples of core assumptions:

Startup Type	Core Assumption
Meal delivery	People will pay a premium for healthy meals they do not cook
AI writing tool	Writers will trust AI to draft initial content
Tutoring marketplace	Parents will book tutors through an app instead of word-of-mouth
B2B analytics	Companies will switch analytics tools for better insights

Your MVP tests this one thing. Nothing else matters yet.

Quick Check: Why should your MVP test your riskiest assumption specifically, rather than building your most requested feature?

Types of MVPs

Not all MVPs require building software:

MVP Type	What It Is	Best For	Example
Concierge	You manually deliver the service	Service businesses	Personally curating meal plans before building an app
Wizard of Oz	Looks automated but humans run the backend	Tech products	A chatbot where you manually respond behind the scenes
Landing page	Describe the product, measure signups	Any product	A page with “Join waitlist” to test demand
Pre-sell	Sell the product before building it	Products with clear deliverables	Crowdfunding campaign or pre-order page
Single-feature	Build one core feature only	Software products	An app that does one thing exceptionally well
Explainer video	Video demonstrating the concept	Complex products	Dropbox famously launched with a demo video

Start with the cheapest option on this list that tests your core assumption. If a landing page can test it, do not build an app.

Building with No-Code and AI Tools

You do not need to be a developer to build an MVP:

No-code platforms:

Bubble: Full web applications without code
Webflow: Professional websites and landing pages
Airtable: Databases and simple workflows
Zapier/Make: Connect tools and automate workflows
Typeform/Tally: Interactive forms and surveys

AI-powered building:

Use AI to generate landing page copy
Use AI to create product descriptions and onboarding flows
Use AI to write help documentation
Use AI to build financial models for pricing experiments

The concierge shortcut: Before automating anything, do it manually. If your product concept is “AI sorts your email,” manually sort 10 customers’ emails first. You will learn infinitely more from doing it by hand than from building an algorithm.

Quick Check: What is a concierge MVP, and why might it be more valuable than building software?

The Feature Decision Framework

For every feature you consider including, ask:

Does this test our core assumption? If no, cut it.
Can customers use the product without this? If yes, cut it.
Will this feature help us learn something important? If no, cut it.
Can we add this later? If yes, cut it for now.

The ruthless rule: If you can describe your MVP in one sentence, it is probably scoped correctly. If you need a paragraph, cut more.

AI-assisted feature scoping:

My startup: [DESCRIBE]
Core assumption: [YOUR RISKIEST ASSUMPTION]

I'm considering these features for my MVP:
[LIST ALL FEATURES YOU ARE THINKING ABOUT]

For each feature, answer:
1. Does it directly test the core assumption?
2. Is it essential for the product to function at all?
3. Can it be added after initial learning?

Recommend the absolute minimum feature set needed to test whether [CORE ASSUMPTION] is true.

Measuring MVP Success

Define success criteria before you launch:

Quantitative metrics:

Sign-up conversion rate (landing page visitors to registered users)
Activation rate (registered users who complete core action)
Retention (users who come back in week 2)
Revenue (if applicable, even small amounts)

Qualitative signals:

Users describing the product to others (word of mouth)
Users asking for features (they care enough to want more)
Users finding workarounds for missing features (high motivation)
Emotional responses (“I have been looking for something like this”)

Set specific thresholds: “If 20% of our beta users return in week 2, we proceed. If less than 5% return, we pivot.”

Try It Yourself

Scope your MVP:

Identify your core assumption in one sentence
Choose an MVP type from the table above
List every feature you think you need, then cut 70% of them
Define three success metrics with specific thresholds
Set a timeline: when will you have something testable?

Key Takeaways

An MVP tests your riskiest assumption with the minimum investment of time and money
Not all MVPs require code: concierge, landing page, and pre-sell models test demand without building software
For every proposed feature, ask whether it tests the core assumption; if not, cut it
Define success metrics and specific thresholds before launching so decisions are evidence-based
If you can describe your MVP in one sentence, the scope is right; if you need a paragraph, cut more

Up Next

In Lesson 5: Crafting Your Pitch Deck, we will translate your validated idea, market research, and MVP results into a pitch deck that makes investors and partners pay attention.