Prioritization
The four-factor scoring model (Reach × Impact × Confidence ÷ Effort) developed by Intercom to compare product and growth bets on a single numeric scale. How it works, when it fails, and our operator-side adjustments.
RICE scores each candidate initiative on four factors and combines them into a single number:
RICE Score = (Reach × Impact × Confidence) ÷ Effort
Higher scores get priority. Lower scores get cut or deferred.
Most prioritization conversations devolve into HiPPO (highest-paid person's opinion) or politics. RICE forces a structured argument: every initiative needs explicit numbers for each factor. The math reveals trade-offs that intuition hides — a high-impact, low-reach initiative often loses to a medium-impact, high-reach one once you do the math.
It also forces explicit confidence. If your team can't agree on whether something is 50% or 80% confidence, that's itself a signal — go research the question before you build.
The framework assumes you can estimate the four factors. For exploratory bets, novel features, or strategic moves with second-order consequences, the estimates are guesses dressed up in numbers.
It also weights short-term thinking. Effort is in the denominator, so multi-quarter foundational work always scores worse than quick wins. Teams that run RICE strictly end up over-investing in optimization and under-investing in platform.
Separate strategic bets from optimization. Don't score them against each other. Run RICE inside the optimization track only. Strategic bets get their own conversation.
Use confidence honestly. A 100% confidence score should be rare. Most well-researched initiatives are 70–80% confidence at best. When everything is 100%, the team is gaming the score.
Re-score quarterly. Initiatives that scored well in Q1 might score differently in Q3 because the world changed. Treat scores as snapshot, not permanent.
ICE Scoring (Sean Ellis): Impact × Confidence × Ease, all on 1–10 scales. Simpler than RICE, useful for high-volume A/B test backlogs.
PIE: Potential × Importance × Ease. Used in CRO contexts (Bryan Eisenberg).
Weighted shortest job first (WSJF): From SAFe (Scaled Agile Framework). Cost of delay ÷ job size.
Kano model: Categorizes features as basic, performance, or delighters. Less score-driven, more strategic.