The Feature That Nobody Used
We built it. 40% of users requested it during discovery. We spent 3 weeks on it.
After launch: 0 users touched it.
The feature was "export to PDF." Users said they needed it. They didn't. They wanted to feel like they had options.
This happens constantly. Here's what we've learned.
The MVP Feature List
Must-Have (Non-Negotiable)
Auth
- Email/password signup
- Password reset
- Email verification (optional for MVP)
That's it. Not OAuth. Not magic links. Not SSO.
Core Feature (ONE Thing)
What's the ONE thing users must be able to do?
Not "task management." But "create tasks and see them in a list."
Narrow it down. Then narrow it more.
Basic Dashboard
Where users go after logging in. Shows their most important data.
Not a full analytics suite. Just: "Here's what's happening."
Settings (Minimal)
Profile picture. Name. Email. Maybe one product-specific setting.
Nothing else.
Should-Have (If Time Permits)
Onboarding Checklist Walk users through the ONE core feature. 5 steps max.
Email Notifications Not 47 emails. Just the critical ones:
- Account created
- Password reset
- Something important happened
Help/Documentation One page. How to do the ONE core thing.
Should NOT Have (Until After Launch)
User Profiles Name and email only. No "bio," "company," "role," "LinkedIn."
Advanced Search Basic text search is fine. Filters can come later.
Bulk Actions Select multiple and... what? Until you know users want this, don't build it.
Export to PDF/CSV See above. Nobody uses these in MVP.
Team Permissions Unless your product is USELESS without roles, skip it.
Custom Themes/Dark Mode Build once. Launch. Add dark mode when users ask.
Mobile App Desktop-first. Always. Unless your product is specifically mobile.
Integrations Stripe. Maybe one other. Not seven.
The Decision Framework
For every feature request, ask:
"If we didn't build this, would users still get value?"
If yes, it's not MVP.
"How many users asked for this?"
One user = ignore. 5 users = consider. 15+ users = maybe. 40%+ users = consider carefully (see: PDF export).
"Does this make the core feature better, or add a new feature?"
Making core feature better = build. Adding new feature = don't build yet.
A Real Example
For a project management MVP:
Build:
- Create task
- Assign due date
- Mark complete
- View all tasks
- Basic search
- Email/password auth
Don't Build:
- Recurring tasks
- Time tracking
- Gantt charts
- Team permissions
- Client portals
- File attachments
- Export to PDF
- Mobile app
- Dark mode
The "don't build" list is longer. That's intentional.
The Real Answer
An MVP should do ONE thing really well.
Not five things poorly.
When users describe your product, they should be able to finish the sentence: "It helps me [ONE SPECIFIC THING]."
If they can't, your MVP has too many features.
What We Tell Founders
When founders send us 47-page requirement docs, we simplify:
"What's the ONE thing users do in your product?"
Usually it's 3-5 sentences.
"That's your MVP scope."
Everything else is "after launch."
This conversation saves months and $50,000+.
Want help scoping your MVP? We help founders focus on what matters.
