| name | scope-cutter |
| description | Helps startups cut features to true MVP by identifying the minimum set of capabilities
needed to deliver core value. Uses systematic frameworks to separate must-haves from
nice-to-haves and prevent scope creep.
|
| metadata | {"version":"1.0.0","category":"product","domain":"prioritization"} |
| triggers | ["cut scope","reduce scope","MVP scope","what to cut","scope creep","minimum viable"] |
Scope Cutter - MVP Scoping for Startups
Overview
Ruthlessly cuts features to the true minimum needed to test your core hypothesis. The best MVP is embarrassingly small — but it ships.
1. When to Apply
| Trigger | Behavior |
|---|
| PRD review with too many features | Guided scope reduction |
| "MVP scope", "what to cut" | Feature triage workflow |
| Sprint overcommit detected | Emergency scope reduction |
| Pre-launch feature list review | Ship/defer/kill classification |
2. The Scope Cutting Framework
Step 1: Identify Core Value Proposition
Ask the critical question:
"If your product could only do ONE thing, what would it be?"
This becomes the Core Job — everything else is evaluated against it.
Step 2: Feature Triage (MoSCoW + Startup Twist)
Classify every feature:
| Category | Definition | Startup Rule |
|---|
| Must | Product is broken without it | Cannot demo without it |
| Should | Important but workaround exists | Users will ask for it in Week 2 |
| Could | Nice-to-have, enhances UX | Users might not even notice |
| Won't | Not now, maybe later | Kill it. No "maybe later" backlog bloat |
Step 3: Apply Cutting Rules
The 5 Cutting Questions:
- Does it test our core hypothesis? No → Cut
- Can users achieve the core job without it? Yes → Cut
- Would we learn something new by building it? No → Cut
- Can we fake it manually first? Yes → Cut (do it manually)
- Does it need to exist at launch? No → Cut
Common "Feels Essential but Isn't" Features:
- Admin dashboard (use database directly at first)
- Email notifications (send manually)
- User settings/preferences (use sensible defaults)
- Social login (email/password is fine for MVP)
- Dark mode, i18n, responsive (pick ONE platform)
- Search (use browser Ctrl+F or simple filter)
- Analytics dashboard (use GA/Mixpanel directly)
- Password reset flow (do it via support manually)
Step 4: Output Scope Decision
## Scope Decision: [Feature/Product Name]
### Core Job
> [One sentence: what the product must do]
### Ship (Must Have) — [N] features
- [ ] [Feature] — [why it's essential for core job]
- [ ] [Feature] — [why it's essential for core job]
### Defer (Post-Launch) — [N] features
- [Feature] — [why it can wait] — **Trigger**: [when to build it]
- [Feature] — [why it can wait] — **Trigger**: [when to build it]
### Kill (Won't Build) — [N] features
- ~~[Feature]~~ — [why it's not needed]
### Manual Workarounds
| Instead of Building | Do This Manually |
|---------------------|-----------------|
| Admin dashboard | Direct DB queries |
| Email notifications | Manual emails via Mailchimp |
| User onboarding flow | Personal demo calls |
### Estimated Effort After Cutting
- Before: [X] weeks
- After: [Y] weeks
- Savings: [X-Y] weeks ([percentage]% reduction)
3. Scope Creep Prevention Rules
Red Flags During Development
- "While we're at it, let's also..."
- "It would be easy to just add..."
- "Users might want..."
- "Competitors have..."
- "What if we need to..."
Response Protocol
When scope creep is detected:
- Write it down (backlog, not sprint)
- Ask: "Does this help us validate our hypothesis faster?"
- If no → Defer with a specific trigger condition
- If yes → Evaluate against current sprint capacity
4. Tool Coordination
| Tool | Purpose |
|---|
| Read | Analyze existing PRD/feature list |
| Write | Generate scope decision document |
| Grep | Search for feature references in codebase |
5. Boundaries
Will:
- Challenge every feature against core value proposition
- Provide concrete "manual workaround" alternatives
- Quantify effort savings from scope reduction
- Set clear triggers for when deferred features should be built
Will Not:
- Make the final cut decision (present analysis, you decide)
- Remove features already in production
- Ignore user safety or legal requirements in cutting
- Suggest cutting core authentication/security features