Skip to content
Back to Idea Validation

Fit Scoring

Score how well your proposed solution fits the problem on 5 dimensions to get an honest go/no-go signal.

Why

Fit scoring is the final gate before you build: a structured, honest assessment of whether your solution actually fits the problem. It scores five dimensions — problem severity, solution effectiveness, reachability, timing, and sustainability — each 1-5, and the total tells you whether to proceed, iterate, or kill. This prevents two failures: building a solution that is technically impressive but does not fit the market (high effectiveness, low severity), and building a solution to a real problem that you cannot sustainably deliver (high severity, low sustainability). The score is not a pass/fail test but a forcing function for honesty: if any dimension is a 1-2, you must address it before building, not hope it resolves.

When

After solution mapping, as the final pre-build gate. Takes 1-2 hours. Re-score after each pivot or major new learning.

How

1. Score problem severity (1-5)
   - How acute is the pain for the target segment?
   - Evidence: interview quotes, willingness to pay, current workaround pain
   - 1 = mild annoyance, 5 = blocks work / costs significant money

2. Score solution effectiveness (1-5)
   - How well does your solution address the pain?
   - Evidence: pretotype results, comparable solutions in adjacent markets
   - 1 = barely helps, 5 = eliminates the pain entirely

3. Score reachability (1-5)
   - Can you reach the target segment cheaply?
   - Evidence: channel exists, audience is findable, CAC is reasonable
   - 1 = no clear channel, 5 = you already have the audience

4. Score timing (1-5)
   - Is the market ready now? (not too early, not too late)
   - Evidence: search trends, competitor activity, enabling tech maturity
   - 1 = way too early / too late, 5 = perfect wave

5. Score sustainability (1-5)
   - Can you deliver this profitably and defend it?
   - Evidence: unit economics, moat potential, operational complexity
   - 1 = unsustainable, 5 = strong margins + defensible

6. Total and decide
   - 20-25: strong fit → build
   - 15-19: conditional fit → address the lowest dimension, then build
   - 10-14: weak fit → iterate on the solution or the segment
   - < 10: poor fit → kill or fundamentally rethink

Example

Solution: AI PR review bot for solo OSS maintainers.

Scores:
  Problem severity:    5  — "I review PRs every weekend, it is destroying my life" (direct quote from 4/15 interviews)
  Solution effectiveness: 4  — pretotype showed 70% of PRs could be pre-vetted; not 5 because complex PRs still need human review
  Reachability:        4  — maintainers are on GitHub/Twitter/Reddit; CAC ~$5 via content marketing; not 5 because no existing audience
  Timing:              4  — LLMs just became good enough for code review in 2024; not 5 because trust is still building
  Sustainability:      3  — $15/mo covers API costs for small repos but margin is thin at scale; moat is per-repo learning data

Total: 20 → strong fit → build

Caveats:
  - Sustainability is the weakest (3): plan to increase price or reduce API cost before scaling
  - Effectiveness is 4 not 5: position as "pre-vet, not replace" to set honest expectations

Decision: proceed to MVP, but address sustainability (test $25/mo price point in pre-orders) before scaling.

Common Problems

  • ⚠Scoring dimensions you have no evidence for — "effectiveness is 4 because I think it will work" is a guess; require evidence (pretotype, comparable) for each score.
  • ⚠Inflating scores to justify building — be honest; a 15 with one dimension at 2 is more useful than a 20 with everything at 4, because it tells you what to fix.
  • ⚠Treating 15 as a pass — 15-19 is conditional; you must address the lowest dimension before building, not proceed and hope it resolves.
  • ⚠Scoring once and never again — fit changes as you learn; re-score after your pretotype, after your first 10 users, after a competitor launches.

Best Practices

  • ✓Require evidence for every score — a score without evidence is a hope; link each to an interview, a test result, or market data.
  • ✓Focus on the lowest dimension, not the total — a 20 with sustainability at 2 will kill you in year 2; fix the weak link first.
  • ✓Re-score at every milestone — the scores change as you learn, and a dimension that was 4 at idea stage may be 2 after a competitor launches.
  • ✓Use the score to decide what to de-risk next — the lowest dimension is your next research priority, not necessarily a reason to kill.