Rank your ideas with RICE

Score each idea by how many people it reaches, how much it helps them, how sure you are, and how much work it takes. The ranking updates as you type. Nothing leaves your browser.

RICE = Reach × Impact × Confidence ÷ EffortHow RICE works ↓

Ranking

Score with your team

1Onboarding checklist1,066.7
2Bulk edit400
3Dark mode250
4Faster search150
5SSO for enterprise37.5

Onboarding checklist leads by 63%. Bulk edit would overtake it with Reach above 2,134, or Effort below 0.7 person-weeks, or Impact 3. That's the estimate worth checking with data before you commit.

Ideas

Reach: People affected per quarter
Impact: 3 massive · 2 high · 1 medium · ½ low · ¼ minimal
Confidence: How sure you are about Reach and Impact
Effort: Total work, in person-weeks

1,066.7

400

250

37.5

150

Saved in this browser only. Use the share link to send it to someone.

How RICE works

RICE asks the same four questions about every idea, so you compare them on evidence instead of on who argues best. It was created by Intercom's product team to order a roadmap full of good ideas (Intercom's original article).

RICE = Reach × Impact × Confidence ÷ Effort

Good for

A backlog of product ideas competing for the same team
Deciding what goes into the next quarter
Making the reasons behind an order visible to everyone

Not for

Bugs and incidents: fix them by severity
Legal or security must-dos
Long bets whose payoff is years away

The four factors

Reach

How many people will this affect in a set time?

Scale: A count of people per quarter. Use the same period for every idea.
Example: About 2,000 new users go through onboarding each quarter: Reach 2,000.
From: Product analytics, sign-ups, support tickets, customer requests.
Watch out: Mixing “per month” for one idea with “per quarter” for another.

Impact

How much will it help each person it reaches?

Scale: 3 massive · 2 high · 1 medium · 0.5 low · 0.25 minimal.
Example: A checklist that gets new users to their first success: 2, high.
From: The one goal you're optimizing for, like activation or retention.
Watch out: Giving everything a 2 or 3. If every idea is high impact, the score stops telling them apart.

Confidence

How much evidence backs your Reach and Impact?

Scale: 100% with data · 80% with some data · 50% mostly a hunch.
Example: The funnel shows where people drop off, but nothing's been tested yet: 80%.
From: Data, research and experiments for these two numbers. Below 50%, go learn more first.
Watch out: Using it for “can we build this?”. Build risk belongs in Effort, not Confidence.

Effort

How much total work will it take?

Scale: Person-weeks or person-months, the same unit for every idea.
Example: Design, build and test the checklist: 3 person-weeks.
From: A rough estimate from everyone involved: design, engineering, QA, content.
Watch out: Counting only engineering time. Effort is the only factor that lowers the score, so leaving work out inflates it.

A worked example

The “Onboarding checklist” from the examples above reaches 2,000 people a quarter, has high impact, some data behind it, and takes 3 person-weeks:

2,000 × 2 × 0.8 ÷ 3 = 1,066.7

“Bulk edit” reaches fewer people with less impact: 800 × 1 × 1 ÷ 2 = 400. The checklist ranks higher, but only as long as its estimates hold. The insight above the ideas shows how much Bulk edit's Reach, Effort or Impact would have to change to overtake it. Those are the numbers worth checking with real data.

Common mistakes

Different time periods: Reach per month for one idea and per quarter for another makes the scores incomparable.
Impact inflation: When most ideas get 2 or 3, Impact stops separating them. Save 3 for the rare idea that changes a metric on its own.
Only counting engineering: Leaving out design, research or rollout makes some ideas look cheaper than they are.
Comparing across teams: Two teams estimate differently. Compare scores within one backlog, not between them.
Treating the score as the decision: The score shows which estimate matters most. Check that one before committing; it's what the insight above the ideas points to.