What is an MVP?
An MVP (Minimum Viable Product) is the simplest version of your product that lets you test your core hypothesis with real users. It's not about launching something bad—it's about learning fast.
Building the Minimum Viable Product
Eric Ries • Lean Startup
Why watch this: Eric Ries literally wrote the book on MVPs. Hear directly from the creator of the Lean Startup methodology about how to build and learn from your MVP.
MVP Purpose
The main purpose of an MVP is to:
Types of MVPs
🎬 Explainer Video
Show what you'll build before building it. Dropbox did this.
📄 Landing Page
A page that describes your product and collects sign-ups.
🧙 Wizard of Oz
Looks automated, but humans do the work behind the scenes.
🤵 Concierge
Manually deliver the service before building technology.
🏗️ Piecemeal
Combine existing tools (Zapier, Airtable, etc.) to create your product.
📱 Single-Feature
Build just one core feature, nothing else.
The Build-Measure-Learn Loop
Build: Create the simplest thing that tests your hypothesis
Measure: Collect data on how users interact with it
Learn: Decide whether to pivot, persevere, or iterate
Match the MVP Types
Match each MVP type to its approach:
Choose Your MVP
What's the best MVP approach?
🌟 Perfect MVP Thinking!
A concierge MVP! Google Form + manual matching lets you validate: Do students sign up? Do mentors participate? Is the matching valuable? You learn in days, not months.
👍 Works, But Slower
Hiring a developer is expensive and slow. You might build the wrong thing. Test manually first to know WHAT to build.
💡 Don't Do This!
6 months to learn to code means 6 months of assumptions untested. You might build something nobody wants. Test the concept first, learn to code later.
Timed Decision: Feature Crisis
You're 2 days from MVP launch. Your co-founder says users in testing want a chat feature. Building it would delay launch by 2 weeks. You have limited runway. Some users said chat is "essential," others didn't mention it. What do you do?
What to Include in Your MVP
Ask yourself: What's the ONE thing that must work for users to get value?
- ✅ Include: Core value proposition (the main thing)
- ✅ Include: Way to measure success (sign-ups, usage, feedback)
- ❌ Exclude: Nice-to-have features
- ❌ Exclude: Perfect design
- ❌ Exclude: Scalability (you'll rebuild later anyway)
Choose Your MVP Type
DecisionDecide what type of MVP makes sense for your idea:
Example MVP Plan
Solution: Platform connecting first-gen students with alumni mentors
Core Hypothesis: First-gen students will actually follow through on mentor meetings (not just sign up and ghost)
MVP Type: Concierge + Piecemeal
Why: I can manually match and check in, while using Calendly + Slack to manage communications
What I'll Build: Typeform for intake, Airtable for matching, Calendly for scheduling, Slack for community
Success Metrics: 20+ students sign up, 10+ actually have mentor meetings, 8+ give positive feedback
Timeline: Launch this weekend!
Cut It Down
FocusList everything you'd want in your product, then ruthlessly cut to MVP essentials:
Example Feature Cut
Full Version Features:
1. AI-powered mentor matching 2. Video chat 3. Resource library 4. Resume review 5. Interview prep 6. Job board 7. Community forum 8. Progress tracking
MVP Core Feature: Mentor matching only
Why: If students and mentors don't connect and find value in conversations, nothing else matters. Fancy features won't fix a broken core.
Cutting: AI matching → manual works fine. Video → use Zoom. Resources → link to existing ones. Resume → maybe phase 2. Rest → later.
🎯 Key Takeaways
- MVP = simplest thing that tests your core hypothesis
- The purpose is learning, not launching a perfect product
- Start manual (concierge), then automate what works
- Include only the ONE feature that delivers core value
- Build-Measure-Learn: iterate based on real feedback