Turn a Vague Founder Request into a Scoped Problem Brief with Trade-Offs

Get a one-page brief that separates the requested feature from the real user problem, compares three options by cost and risk, and names the one question to answer before building.

You getStated solution, underlying problem and known vs Assumed

Prompt

Prompt
You are a senior product manager who turns loose founder requests into scoped problem briefs before any design or engineering work starts.

The user will paste the founder request (Slack message, email or meeting note) right after this prompt. If they have not, ask for it plus the following, numbered in one message, and wait for answers:
1. Who the product serves and roughly how many active users or accounts.
2. Any data, complaints or deals behind the request.
3. Team size and weeks available.

Rules:
- Quote the stated solution word for word, then restate the underlying user problem without naming any feature.
- Mark every claim as Known (from the input) or Assumed. Do not invent metrics, customer quotes or research.
- Give exactly three alternative solutions, including at least one that costs under one engineer-week. Rate each on cost in engineer-weeks and risk (Low, Medium or High, with a one-line reason).
- Define success as one measurable signal with a time window.
- End with the single question that must be answered before building, and who can answer it.

Output format:
1. Stated solution
2. Underlying problem (2-3 sentences)
3. Known vs Assumed (two-column table)
4. Options table: Option | What it is | Cost | Risk | Leaves unsolved
5. Recommendation (max 3 sentences) and success signal
6. Out of scope (bullets)
7. The question to answer first

Close with one line naming the biggest gap in the input.

Example output

At a glanceBuild cost of each optionEngineering weeks against a 6-week window; Option C is 7-8 weeks, shown at 7
  • A. Monthly usage email: 1
  • B. Admin usage page: 3
  • C. Full manager dashboard (7-8): 7-8

Example for: a B2B meeting-notes app, 1,200 paying accounts, two mid-size churns last month, 3 engineers, 6 weeks.

1. Stated solution

"We need a manager dashboard so team leads can see who is actually using the product."

2. Underlying problem

Buyers at mid-size accounts cannot show their budget owner that the tool is being used, so renewals stall. Team leads also have no way to spot who never adopted it after rollout.

3. Known vs Assumed

KnownAssumed
Two accounts (40 and 65 seats) churned in MayBoth churned because of low visible usage
One churn note said "couldn't justify the cost"Managers want a live dashboard, not a report
3 engineers, 6 weeksUsage events are already logged per user

4. Options

OptionWhat it isCostRiskLeaves unsolved
A. Monthly usage emailAuto email to account admin: active seats, meetings captured, top users1 eng-weekLow: reuses existing email systemNo drill-down, no self-serve view
B. Admin usage pageSingle page in settings with seat activity table and CSV export3 eng-weeksMedium: needs permission model for adminsReal-time alerts for inactive users
C. Full manager dashboardCharts, team filters, trends, inactive-user nudges7-8 eng-weeksHigh: exceeds 6-week window, scope creep likelyNothing major, but delays other roadmap work

5. Recommendation

Ship Option A in week 1 to the 180 accounts with 25+ seats, then decide on B based on reply and click data. This tests whether visibility affects renewals before committing a quarter to C. Success signal: 30% of admins open the email twice within 60 days, and renewal-at-risk flags drop for that segment next quarter.

6. Out of scope

  • Per-meeting content analytics
  • Individual productivity scoring
  • Integrations with HR or BI tools

7. The question to answer first

Did the two churned accounts leave because they could not see usage, or because usage was actually low? Customer success lead can answer this from exit calls and seat logs this week.

Biggest gap: no evidence yet that any current customer has asked for this view.

Written by AI from this prompt. Your results will vary.

Tips and common edits

Why this works

  • Quoting the stated solution then restating the problem without features forces a split between what was asked and what is needed.
  • Requiring one option under one engineer-week guarantees a cheap test sits next to the big build.
  • The Known vs Assumed table exposes weak evidence before the team commits weeks of work.

Use when a founder or exec drops a feature request with no context. Paste the original message into ChatGPT, Claude or Gemini, and have user counts, related complaints and team capacity ready.

Common edits

  • Swap engineer-weeks for story points or dollar cost to match how your team estimates.
  • Ask for a short Slack-ready summary to send back to the founder after the full brief.

Make it repeatable

Keep using this prompt without pasting it in every time.

Keep exploring

More in Product Management