---
title: "RICE Score Calculator: Copy the Ranked List | Sleekplan"
canonical_url: "https://sleekplan.com/tools/rice-score-calculator"
last_updated: "2026-09-19T23:33:34.615Z"
meta:
  description: "RICE is Reach times Impact times Confidence, divided by Effort. Score each feature on Intercom's 3 to 0.25 impact scale, then copy the ranked table."
  "og:description": "RICE is Reach times Impact times Confidence, divided by Effort. Score each feature on Intercom's 3 to 0.25 impact scale, then copy the ranked table."
  "og:title": "RICE Score Calculator: Copy the Ranked List | Sleekplan"
---

# A RICE score calculator that ranks your backlog and copies out as a table.

### Scoring table

#### RICE scoring table

Reach is customers per quarter. Impact is 3, 2, 1, 0.5 or 0.25. Confidence is 100%, 80% or 50%. Effort is person-months, minimum 0.5. RICE is (Reach x Impact x Confidence) / Effort, rounded to one decimal.

##### Ranked backlog

| Rank | Feature | Reach | Impact | Confidence | Effort | RICE |

| --- | --- | --- | --- | --- | --- | --- |

| 1 | Bulk invite teammates by CSV | 900 | 1 | 100% | 1 | 900.0 |

| 2 | Saved filters on the board | 1400 | 0.5 | 100% | 1.5 | 466.7 |

| 3 | Slack notification on status change | 620 | 1 | 80% | 2 | 248.0 |

| 4 | Merge duplicate requests | 300 | 2 | 50% | 1.5 | 200.0 |

| 5 | Dark mode | 2000 | 0.25 | 100% | 3 | 166.7 |

| 6 | Single sign-on with Okta | 120 | 3 | 80% | 4 | 72.0 |

##### Scale

Impact: 3 massive, 2 high, 1 medium, 0.5 low, 0.25 minimal. Confidence: 100% high, 80% medium, 50% low. Both scales are fixed; they are not free numbers.

 Markdown. Paste it straight into your docs, tracker or a coding agent.

### Quarterly planning

#### RICE ranking for [quarter]

Scored on [date] by [names]. Reach is customers per quarter, Effort is person-months. Anything below the cut line is ranked, not rejected.

##### Committed

| Rank | Feature | Owner | Carried over | Reach | Impact | Confidence | Effort | RICE |

| --- | --- | --- | --- | --- | --- | --- | --- | --- |

| 1 | Bulk invite teammates by CSV | [name] | no | 900 | 1 | 100% | 1 | 900.0 |

| 2 | Saved filters on the board | [name] | Q2 | 1400 | 0.5 | 100% | 1.5 | 466.7 |

| 3 | Slack notification on status change | [name] | no | 620 | 1 | 80% | 2 | 248.0 |

##### Below the cut line

| Rank | Feature | Reach | Impact | Confidence | Effort | RICE | Revisit when |

| --- | --- | --- | --- | --- | --- | --- | --- |

| 4 | Merge duplicate requests | 300 | 2 | 50% | 1.5 | 200.0 | Confidence rises above 50% |

| 5 | Single sign-on with Okta | 120 | 3 | 80% | 4 | 72.0 | Reach passes 400 accounts |

##### Capacity

Person-months available this quarter: [number]. Person-months committed above: [number]. If the second number is larger, move the cut line, do not shrink the estimates.

 Markdown. Paste it straight into your docs, tracker or a coding agent.

### Spreadsheet formula

#### RICE in Sheets or Excel

Six columns and one formula. Identical in Google Sheets and Excel.

##### Columns

A: Feature. B: Reach. C: Impact. D: Confidence. E: Effort. F: RICE. G: Rank.

##### The formula in F2, filled down

=ROUND((B2*C2*D2)/MAX(E2,0.5),1)

##### Why MAX(E2,0.5)

It is the divide-by-zero guard. An empty or zero Effort cell would return #DIV/0! or an infinite score that sorts to the top of your roadmap. 0.5 person-months is the smallest honest estimate anyway.

##### Rank in G2, filled down

=RANK(F2,$F$2:$F$50)

##### Data validation for column C (Impact)

3, 2, 1, 0.5, 0.25 as a dropdown list. Labels: 3 massive, 2 high, 1 medium, 0.5 low, 0.25 minimal.

##### Data validation for column D (Confidence)

1, 0.8, 0.5 as a dropdown list. Labels: 1 high (100%), 0.8 medium (80%), 0.5 low (50%). Store the decimal, format the cell as a percentage.

##### Then

Sort by column F, Z to A. Freeze row 1. Do not let anyone widen the Impact or Confidence dropdowns.

 Markdown. Paste it straight into your docs, tracker or a coding agent.

### With customer evidence

#### RICE with the Reach evidence attached

One extra column, and it is the one that ends the argument. Reach is the only number in RICE that can be invented, so write down where each one came from while you still remember.

##### Ranked backlog

| Rank | Feature | Reach | Where Reach came from | Impact | Confidence | Effort | RICE |

| --- | --- | --- | --- | --- | --- | --- | --- |

| 1 | Bulk invite teammates by CSV | 900 | Workspaces that added 5 or more members last quarter, billing export | 1 | 100% | 1 | 900.0 |

| 2 | Saved filters on the board | 1400 | Accounts that opened the board at least once last quarter, analytics | 0.5 | 100% | 1.5 | 466.7 |

| 3 | Slack notification on status change | 620 | Workspaces with Slack connected, integrations table | 1 | 80% | 2 | 248.0 |

| 4 | Merge duplicate requests | 300 | Admins who merged at least one post last quarter | 2 | 50% | 1.5 | 200.0 |

| 5 | Single sign-on with Okta | 120 | Accounts on a plan that allows SSO, 14 of which asked in sales calls | 3 | 80% | 4 | 72.0 |

##### Rule for the fourth column

A count and its source, or the row does not get scored. "Customers have asked for this" is not a Reach number. "41 votes across 3 merged posts, 9 of them on accounts above 50 seats" is.

 Markdown. Paste it straight into your docs, tracker or a coding agent.

The calculator loads here: score several features at once, and copy the ranked list as a markdown table. RICE = (Reach x Impact x Confidence) / Effort, with Impact from 3, 2, 1, 0.5, 0.25 and Confidence from 100%, 80%, 50%. The formula, both scales and the worked example on this page are readable without it.

## The RICE score formula

RICE = (Reach x Impact x Confidence) / Effort. Reach is the number of people a change affects in one fixed period, counted in real units such as customers per quarter. Impact is a multiplier picked from a fixed five-value scale. Confidence is a percentage picked from a fixed three-value scale, and it enters the arithmetic as a decimal, so 80% is 0.8. Effort is what the change costs, counted in person-months, where one person-month is the work one team member does in one month.

Sean McBride, who published the framework on the Intercom blog on 5 January 2018, describes the result as total impact per time worked. That is the whole idea in five words. The numerator is how much good the change does and how sure you are of it, the denominator is what it costs to find out, and dividing one by the other is what lets a small change that touches everyone outrank a large change that three people asked for.

The score has no unit of its own, which means it does not travel. Effort is the live disagreement: the framework says person-months, but published RICE calculators variously use person-weeks or story points, and the same feature scores roughly four times higher in weeks than in months. Fix the unit before you score the first item and write it at the top of the table, because a RICE number with no stated Effort unit cannot be checked by anyone.

## Reach, Impact, Confidence and Effort

Reach is a count of people or events in one time period. Not a percentage, not a feeling. Pick the period, use the same one for every item on the list, and take the number from product metrics rather than from a hat. The original worked example: 500 customers reach this point in the signup funnel each month and 30% choose this option, so the reach is 500 x 30% x 3 = 450 customers per quarter.

Impact is picked from exactly five values: 3 for massive, 2 for high, 1 for medium, 0.5 for low and 0.25 for minimal. It is a per-person multiplier, not a total, and it is scored against one named goal rather than against the general good of the product.

Confidence is picked from exactly three values: 100% for high, 80% for medium and 50% for low. It covers the other three numbers, not the idea. The source is blunt about what sits underneath: anything below 50% is a total moonshot, which is a decision to make on other grounds rather than a number to score.

Effort is estimated in person-months, all disciplines included, kept to whole numbers, with 0.5 for anything well under a month. Rough is the intent. An Effort estimate carried to one decimal place implies a precision that the other three numbers cannot support.

Impact and Confidence are fixed scales, and that is the part worth defending. Reach and Effort are measurable, so someone who disagrees with you can go and check them. Impact and Confidence are judgements, and a judgement you can set to any number is a dial you turn until the thing you already wanted comes out on top. Five values and three values force the judgement into the open: you have to say this one is high and that one is medium, out loud, in front of the people who will build both. A number anyone can set to anything ranks nothing.

One correction worth carrying, because the mistake is now widespread. The Impact scale runs from 3 down to 0.25, not from 3 down to 0.5. Google's AI summary for this framework has stated the range as 0.5 minimal to 3.0 massive, which deletes the bottom value and relabels the one above it. In the published scale 0.5 is low and 0.25 is minimal, and the difference matters: it is the gap between a feature scoring half and a quarter of medium.

## A worked RICE score, start to finish

Take saved filters on a feedback board. Analytics say 1,400 accounts opened the board at least once last quarter, so Reach is 1,400. It saves a few seconds per visit rather than unlocking anything new, so Impact is 0.5, low. The reach comes from measured behaviour rather than from a survey, so Confidence is 100%. Two engineers think it is about six person-weeks between them, so Effort is 1.5 person-months.

1,400 x 0.5 = 700. 700 x 1.0 = 700. 700 / 1.5 = 466.7.

Now single sign-on with Okta, on the same list. Only 120 accounts are on a plan that would use it, so Reach is 120. For those accounts it removes a procurement blocker, which is as big as it gets, so Impact is 3. Sales have asked for it repeatedly but nobody has measured how many accounts would switch it on, so Confidence is 80%. It is four person-months of work.

120 x 3 = 360. 360 x 0.8 = 288. 288 / 4 = 72.0.

466.7 against 72.0 is the argument the framework exists to have. Single sign-on is the bigger piece of work and that is precisely why it loses: Effort is the denominator. Saved filters wins this quarter not because it matters more to any one customer, it matters far less, but because it touches twelve times as many accounts for a quarter of the time. And if you think the ranking is wrong, the framework has already told you where to argue: there are eight numbers behind those two scores and one of them is the one you disagree with.

## Which RICE prioritization template to use

All four templates above score the same way. What changes is what sits next to the score, and who reads it afterwards.

Pick by what happens to the output. Ranking a backlog once, in a meeting, needs nothing more than the scoring table. If the ranking becomes the plan of record, the quarterly variant carries the owner, the target quarter and the cut line, so the ranking and the commitment stay in one document. If you want the arithmetic to keep working after somebody adds a row next month, the spreadsheet variant gives you the cell formula and the two dropdown lists. And if anyone is going to ask where a Reach number came from, use the evidence variant from the start: adding the sources afterwards turns a scoring session into an archaeology session.

Which RICE template to use when

| Template | Use it when | Rows it suits | What it does not do |
| --- | --- | --- | --- |
| Scoring table | You are ranking a backlog once, in a meeting | 5 to 15 | No owner, no dates, no capacity check |
| Quarterly planning | The ranking becomes the plan of record | 10 to 30 | Heavier to fill in, and it goes stale |
| Spreadsheet formula | You want the arithmetic to live in Sheets or Excel | Any number of rows | Lives outside the doc everyone reads |
| With customer evidence | Someone will ask where Reach came from | 5 to 15 | Needs the evidence to exist first |

## What a RICE score is worth once you have it

Higher is better, and that is close to everything a single RICE score tells you. There is no universal threshold. Absolute cutoffs are published, and one widely linked calculator asserts that anything above 100 warrants prioritization while anything below 20 belongs in a someday pile, but those numbers are a property of that team's reach counts rather than of the framework.

The reason is arithmetic. Reach is an absolute headcount, so a team with 40,000 users generates scores two orders of magnitude larger than a team with 400 for identical work. Effort is a local estimate in a locally chosen unit. A RICE score is therefore only comparable inside one backlog, scored by one team, in one sitting. Carried across teams, or across a quarter in which the reach baseline moved, it compares nothing.

Which is why the ranking is the output and the individual score is an intermediate. The useful question is never whether 340 is good, it is what 340 beat and what beat it. If the top of your list is obviously wrong, one of the four numbers behind it is wrong, and RICE has done its job by telling you which row to open.

RICE also answers a narrower question than it first appears. It ranks candidates you have already decided are worth considering. Deciding what is in scope at all, and what is explicitly not, is a different job with a different tool.

[Compare it with a MoSCoW list ](https://sleekplan.com/tools/moscow-prioritization-template)

## Where the Reach number comes from

Almost every real objection to RICE is an objection to Reach. Impact and Confidence come from fixed scales, so the argument about them is short and public. Effort comes from the people who will do the work, who are in the room. Reach is the one number that can be invented, and an invented Reach makes the whole ranking decorative.

The instruction in the original is to use real measurements from product metrics instead of pulling numbers from a hat. In practice that means one of three things: a count from analytics, how many accounts touch this surface in a quarter; a count from demand, how many people asked for it and how many voted for the same thing; or a segment, which plan or which size of account the number is true for.

The middle one is the one most teams cannot produce on the spot, because the requests are scattered across support tickets, sales calls and a Slack channel somebody muted. That is the part Sleekplan holds. Feedback from the in-app widget, the public portal and your support tools lands on one board, duplicates merge into a single post that keeps every voter, and each post carries its vote count and the segment those voters belong to. Score in whatever table you like: what changes is that the number in the Reach column has a link behind it, and the roadmap it feeds tells the people who voted what was ranked above them and what shipped.

![A Sleekplan feedback board with one request open beside it, its engagement rail showing 621 upvotes, 133 downvotes, 96 comments and 21 subscribers, and five similar posts listed below, the kind of count that belongs in the Reach column of a RICE prioritization score](https://sleekplan.com/_ipx/f_webp&s_1680x960/images/tools/rice-score-calculator/featured-request-vote-counts-2x.webp)_Reach is the one RICE number people invent. This is the same number with a list of names behind it._

[See how the roadmap works ](https://sleekplan.com/roadmap)

## How to calculate a RICE score

### Count the Reach in one fixed period

Decide the period first, then use it for every item on the list: customers per quarter, transactions per month, signups per week. Reach is a count of people or events, never a percentage. Take it from product metrics rather than from a hat, and if you have to combine numbers, show the working: 500 customers hit this step each month and 30% pick this option, so 500 x 30% x 3 = 450 per quarter.

### Pick the Impact multiplier from the five values

3 for massive, 2 for high, 1 for medium, 0.5 for low, 0.25 for minimal. Five values, no others, and no typing a number of your own. Score it per person reached, against one named goal, not against how important the project feels. If two items both look like a 3, one of them is a 2 and the conversation about which is the useful part of the exercise.

### Set Confidence from the evidence you actually have

100% when the reach and impact numbers come from measured behaviour, 80% when you have decent data for some of it, 50% when the whole thing rests on a strong hunch. Confidence is about your other three numbers, not about whether the idea is good. Anything you would score below 50% is a moonshot: decide it on other grounds instead of dressing it up with a score.

### Estimate Effort in person-months

One person-month is the work one team member does in a month, counting every discipline the change needs: design, engineering, QA, docs, migration. Stick to whole numbers, with 0.5 for anything well under a month, and never use zero. Rough is deliberate here. The other three numbers cannot support a precise denominator, and an Effort of 2.75 implies a certainty nobody in the room has.

### Divide, then rank the whole list

(Reach x Impact x Confidence) / Effort, rounded to one decimal, with Confidence as a decimal so 80% is 0.8. Then sort the list descending and read the order, not the numbers. The rank is the output. A single score has no natural unit and no threshold, so it means nothing until it sits above or below another score produced by the same team in the same sitting.

FAQ

## Questions people ask

### How do you calculate a RICE score?

Multiply Reach by Impact by Confidence, then divide by Effort. Reach is a count of people or events in one fixed period, Effort is person-months, and Confidence enters as a decimal, so 80% is 0.8. A feature reaching 620 accounts a quarter at medium impact, 80% confidence and two person-months scores (620 x 1 x 0.8) / 2 = 248.0. The result has no natural unit, so it only means something next to another score from the same list.

### What is the RICE formula?

RICE = (Reach x Impact x Confidence) / Effort. The letters stand for Reach, Impact, Confidence and Effort. Reach and Effort are measured in real units, people per period and person-months. Impact and Confidence are picked from fixed scales rather than typed. Sean McBride published the formula on the Intercom blog on 5 January 2018 and describes what it produces as total impact per time worked.

### What is the RICE method of prioritization?

A way of ranking candidate work by scoring each one on four factors and dividing the benefit by the cost. Two of the four are deliberately not free numbers: Impact is picked from 3, 2, 1, 0.5 or 0.25, and Confidence from 100%, 80% or 50%. Constraining them is the method, not a simplification of it, because an impact figure anyone can set to anything makes the ranking unfalsifiable. Intercom published it in 2018 to separate ideas its own team could not otherwise rank.

### Is a higher or lower RICE score better?

Higher is better, but only within one list. Reach is an absolute headcount and Effort is a local estimate, so a team with 40,000 users produces much larger scores than a team with 400 for exactly the same work. A score is comparable to the other scores in the same backlog, scored by the same people in the same sitting, and to nothing else. Comparing your 340 with a number from another company's blog post compares nothing.

### How do you interpret a RICE score?

Read the order, not the number. There is no universal cutoff, and published thresholds such as above 100 is a priority or below 20 is a someday reflect one team's reach counts rather than anything about the framework. Ask what the score beat and what beat it. When the top of the list looks wrong, one of the four inputs behind it is wrong, and finding which one is the point of scoring in the first place.

Get started in 4 minutes

## **Your roadmap,**** on **autopilot**. **

Free for 30 days. No card, no sales call. Cancel anytime.

[**Start free **](https://app.sleekplan.com/sign-up/) [**Book a 20-min demo **](https://sleekplan.com/contact/sales)