PlayInfoBox

Home /Utilities /RICE Scoring Explained: Reach, Impact, Confidence, Effort

RICE Scoring Explained: Reach, Impact, Confidence, Effort

RICE turns "which project should we do first" into a number -- here is what actually goes into that number.

Reach: how many people does this affect in a given period

Reach is a concrete count, not a vague sense of scale -- typically something like "customers affected per quarter" or "users who will see this feature per month." It should be an estimate you could defend with a rough calculation, not a gut feeling.

Impact: how much does it move the needle per person affected

Impact is usually scored on a simple scale (for example: 3 = massive, 2 = high, 1 = medium, 0.5 = low, 0.25 = minimal) rather than measured precisely, since the actual effect on an individual user is rarely knowable in advance with real precision.

Confidence: how much you trust your own reach and impact numbers

Confidence, expressed as a percentage, is a deliberate discount applied to optimistic guesses -- 100% for numbers backed by solid data, 80% for a reasonable estimate, 50% for a real guess. It exists specifically to stop a confidently-stated but weakly-supported number from outscoring a more honestly-uncertain one.

Effort: total person-time required, across every role involved

Effort is usually measured in person-months and should include everyone's time -- engineering, design, QA, and any other role -- not just the most visible contributor's time, since underestimating effort is one of the most common ways a RICE score ends up misleading.

Putting it together: (Reach x Impact x Confidence) / Effort

The final score divides the combined benefit (reach times impact times confidence) by the cost (effort), so a higher score means more estimated benefit per unit of work required -- useful for ranking a list of candidate projects against each other, not for evaluating one project in isolation.

What RICE is actually good for

RICE is best used to rank a list of comparable candidate projects against each other using a shared, semi-objective method, replacing pure gut-feel or whoever argues loudest in the room. It works best when the inputs are genuinely estimated with some rigor rather than reverse-engineered to justify a decision someone already wanted to make.

Where RICE scores tend to mislead

Because Impact and Confidence are inherently subjective scales, two people scoring the same project can land on meaningfully different numbers, and a low-effort, low-reach project can outscore a strategically important but resource-intensive one. RICE should inform a prioritization conversation, not replace it entirely.

Frequently Asked Questions

What counts as a good RICE score?

There is no universal threshold -- RICE scores are only meaningful relative to each other within the same batch of candidate projects scored by the same team using the same scales, not as an absolute number compared across different contexts.

Should Confidence ever be 100%?

Rarely, and it's worth being suspicious of a project consistently scored at 100% confidence, since that implies the reach and impact numbers are backed by data as solid as a completed measurement, which is uncommon for a project that hasn't shipped yet.