RICE Scoring and MVP Synthesis

Purpose

Turn a comparison of multiple solution alternatives into an objective, justified MVP choice using the RICE model, and translate that choice directly into a usable strategy: what the MVP exactly does, how it's positioned in one sentence, and why exactly this one wins. This is a bridge skill between solution ideation (many alternatives) and writing the PRD (one chosen direction).

Based on

• The RICE prioritization model (Reach, Impact, Confidence, Effort) — a generally known product-prioritization framework, not the owner's own.
• The methodology of an external "AI-first SaaS Product" workshop, applied by the owner to one own case — see ../../references/ai-first-saas-workshop-source.md and the worked example ../../cases/ai-decision-coach-mvp-case.md, sections 6–7. Note: applied only once so far — not broadly validated.

Method (draft — to be filled in further)

1. Score each solution direction on four criteria (1–5):
   • Reach — how many users this would touch.
   • Impact — how significant the effect is for the user it touches.
   • Confidence — how sure you are that your Reach/Impact/Effort estimates hold up (high = strong evidence, low = a guess).
   • Effort (inverted: 5 = easiest/lowest effort, 1 = hardest/highest effort) — note the inversion: in this model a high Effort score means LOW build cost, so all four criteria sum in the same direction (higher = better MVP candidate).
2. Calculate the total RICE score (max 20 across the four criteria, or scale as needed) for each direction and rank them.
3. Briefly justify the score for each criterion — don't leave scores unexplained. In particular, tie the Effort estimate concretely to the existing tech stack/data/tools (what already exists vs. what needs to be built from scratch).
4. Choose the MVP with the highest RICE score, UNLESS a specific strategic reason favors another (e.g. a higher-Effort direction is the only one that proves a genuine differentiator, not just table-stake value). If you deviate from the highest score, explicitly justify why.
5. Write the MVP definition (2-3 sentences). Combine the chosen AI wedge (differentiator need) and the essential table-stake needs into one concise description of what the MVP does and for whom.
6. Sketch the MVP flow concisely (5-8 steps): user input → AI synthesis/scoring → decision engine/logic → AI output(s) → next-step plan → (optional) communication support → (optional) path to deeper tools.
7. Write a one-sentence positioning statement. Format: "[Product] gives [target customer] [core benefit] through [distinctive mechanism]." Test: could this sentence describe any competitor? If yes, it isn't specific enough yet.
8. Write 3 "why we win" claims. Each claim ties one strength (differentiator need, existing data/tool, unique approach) to a concrete competitive advantage — not general claims ("we're better") but justified reasons.
9. Carry the MVP definition, flow, positioning statement, and "why we win" claims into ../ai-buildable-prd-writing/SKILL.md as the basis for the PRD.

What this skill does NOT do

• Does not make the final MVP choice for you in a fully mechanical way — the RICE score supports the decision, it isn't an automatic rule; a strategic deviation from the highest score is allowed if justified.
• Does not assess financial viability or unit economics — only relative prioritization between solution alternatives. See ../../../../business-case-and-analysis/skills/roi-npv-sensitivity-model/SKILL.md for deeper financial modeling once the MVP is chosen.
• Does not replace ../../../opportunity-recognition/skills/opportunity-evaluation- and-judgment/SKILL.md — this is a narrower, faster choice between 2-3 already-identified solution directions, not a full assessment of an opportunity from scratch.

Refinement notes

This skill has so far been applied to one case (the owner's own case). As you apply it to more businesses, add:

• your own rules of thumb for when it's worth deviating from the highest RICE score
• concrete examples of positioning statements and "why we win" claims from other cases in the ../../cases/ folder

Once this section has been filled in with multiple cases, raise skills_index.json's maturity field to validated (see ../../../../meta/maturity_levels.md).

Continue from here

• Preceding skill in this pack: ../ai-differentiator-solution-ideation/SKILL.md — produces the three alternatives scored here.
• Next skill in this pack: ../tiny-core-identification-and-feature-freeze/SKILL.md — once a direction is chosen here, isolates the one interaction that is its actual reason to exist and blocks feature creep before the PRD is written.
• Next skill in this pack: ../ai-buildable-prd-writing/SKILL.md — writes the PRD for the chosen MVP.
• Related skill in this pack, if the chosen MVP is a conversational/ agentic product: ../ai-native-conversational-os-design/SKILL.md.
• Worked example: ../../cases/ai-decision-coach-mvp-case.md, sections 6–7.
• The pack's shared guardrails: ../../CLAUDE.md

References

• ../../references/ai-first-saas-workshop-source.md — source information
• ../../cases/ai-decision-coach-mvp-case.md — worked example
• ../../CLAUDE.md — the pack's shared guardrails