RICE Score Calculator
Score a backlog on reach, impact, confidence and effort, then see which items are quick wins and which one deserves a second look.
What this calculates
This multiplies reach by impact by confidence, divides by effort, and ranks the result. The output is a defensible ordering rather than a decision, because every input is an estimate and the score inherits all four of them. Its real value is that it makes the estimates visible enough to argue with.
Scoring a four-item B2B SaaS backlog with RICE, self-serve onboarding ranks first at 1,440, which is 50% clear of the second item, and two of the four items fall in the quick-wins quadrant of high value and low effort.
Your backlog
| # | Feature ↕ | Reach ↕ | Impact ↕ | Confidence ↕ | Effort ↕ | Score ↓ | |
|---|---|---|---|---|---|---|---|
| 1 | 1,440 | ||||||
| 2 | 960 | ||||||
| 3 | 533 | ||||||
| 4 | 31 |
Effort
Build this first
LiveSelf-serve onboarding
Score 1,440, 50% clear of second place
- Items scored
- 4
- In quick wins
- 2
- Low-confidence rows
- 1
One thing to check
Admin audit log carries the largest effort estimate and the lowest confidence in this list, which is the combination that produces the worst surprises. Before it moves up the order, either shrink the scope until the effort estimate is one somebody can defend, or spend a week raising the confidence.
Pressure-test this against your real numbers
Thirty minutes on the estimates behind the top three, because that is where the ordering actually comes from.
No email required to see your result. The field above exists only if you want a copy.
How is this calculated?
Reach multiplied by impact multiplied by confidence, divided by effort. Reach and effort carry units you choose and keep consistent across the list. Impact and confidence are deliberately coarse scales, because a finer scale invites false precision in a number that is a judgment call either way. The quadrant plots the numerator against effort, so the top-left cell is high value for low effort.
Formula
RICE score = (reach × impact × confidence) ÷ effort Quadrant position uses reach × impact × confidence on the vertical axis and effort on the horizontal.
- Reach is the number of accounts touched in one quarter. Any consistent unit works, but mixing accounts and users across rows silently reweights the whole list.
- Impact runs 0.25 for minimal, 0.5 for low, 1 for medium, 2 for high and 3 for massive. The scale is coarse on purpose.
- Confidence is 0.5 for a hunch, 0.8 for evidence from a handful of customers, and 1.0 for something you have measured.
- Effort is person-months on the same scale for every row: 1 is about a day, 3 is about a week, 5 is over a month.
- The score has no units and is meaningful only against the other rows in the same list. Comparing scores across two different backlogs is not a thing you can do.
When this number misleads
RICE rewards work that is easy to estimate, and the most valuable thing on most roadmaps is the one nobody can size yet. A platform change that unblocks three later features scores badly here because its reach is indirect and its effort is honest. Score it anyway, then look at what sits at the bottom of the list and ask whether it is there because it is unimportant or because it is hard to describe.
Questions founders ask about this
Should the highest score always go first?
No. Use the ordering as the default and then apply the reasons it cannot see: sequencing dependencies, a contractual commitment, a team that can only work on one of these. If you override the order, the score is still useful because it makes the size of the override visible.
Why does confidence divide rather than gate?
Because a low-confidence item is not worthless, it is uncertain. Multiplying by 0.5 halves its position rather than removing it, which leaves it in the list where somebody can see it and decide to go and raise the confidence instead.
How many rows should a backlog have here?
Enough to cover a quarter, which is usually between six and fifteen. Past about twenty the exercise stops being prioritization and becomes inventory, and the scores at the bottom of a long list are never revisited anyway.
Do you store what I enter?
No. The calculation runs in your browser and nothing is transmitted. Your last inputs are saved in your own browser so the page remembers them when you return. If you use the email field, only the result summary and your address are sent.
Book a 30-minute product strategy session
Bring your top three. We will pressure-test the estimates that put them there.
Book a strategy sessionNo pitch deck. No follow-up sequence.