rice-prioritization

Ranks a backlog by RICE = (Reach × Impact × Confidence) / Effort, built at Intercom to settle backlog fights without the loudest voice winning. Every input is a guess, but all items are guessed on the same definitions, so the ranking reflects relative value rather than relative volume of advocacy.

The formula, the four factors, their scales, the built-in Confidence bias check, the sensitivity logic, and the pitfalls live in references/rice.md. Load that file before scoring.

When to use

The user has many roadmap initiatives or features that differ in reach, impact, and effort, and wants them sequenced by expected value. RICE sequences a backlog; it does not pick among a handful of designs. Use weighted-decision-matrix to CHOOSE among a few options on bespoke weighted criteria. Use impact-feasibility-matrix for a quick two-axis sort when reach and confidence don't matter enough to model. Use kano-model to understand WHICH features matter and why; Kano explains demand, RICE sequences supply. If there are only two or three items, the overhead of calibration buys little: rank them by hand.

Language

Conduct the session and write the report in the language the user is writing in. The reference file is English; keep the framework keyed to it regardless of the output language.

Inputs

• INITIATIVES (required): the backlog of features or initiatives to rank. If only a goal is given without a list, ask for the items first.
• GOAL (optional): the metric Impact is measured against (activation, revenue, retention). If absent, fix it with the user in Stage 1; without it Impact has no referent.
• CONTEXT (optional): known data (usage figures, past effort actuals), the planning horizon, and any hard constraints or commitments that pre-empt the ranking.