KjerrThe formula
Reach × Impact × Confidence ÷ Effort, on the scales below. Pick a value in each column and the score updates; hover it and the arithmetic is shown.
RICE prioritization,
with the evidence beside it.
RICE prioritization in Kjerr lives on an internal roadmap the team keeps to itself: every feature request becomes a row, Reach × Impact × Confidence ÷ Effort becomes a score, and the votes, the companies and the MRR behind the request sit in the same row. The public board never shows a score.
What a RICE score is, and what Kjerr adds
A RICE score is four coarse estimates multiplied together, so a team can rank twenty ideas in an afternoon instead of arguing about each one. Kjerr computes it and puts the evidence next to it.
Evidence beside each score
Votes, distinct companies and the MRR those companies pay, read from Users & Companies, so Reach is argued from data rather than guessed.
The line
Give the roadmap an effort budget, say twelve points for the quarter. Sorted by score, the table draws a line where the budget runs out and sums the votes and MRR above it.
The four factors, and the scales Kjerr uses
Coarse scales are the point of RICE scoring: powers of two keep the arguments short and the ranking honest. These are the defaults; every scale can be rewritten under Scoring.
- Reach. Everyone in the product (4), some users, a plan or a region (2), new users only (1). The votes and companies in the row say which.
- Impact. Game changer (4), significant value (2), some value (1), low value (0.5). Judged for the whole product, not for the feature alone.
- Confidence. High, 80%: plenty of feedback and the numbers agree. Medium, 50%: plenty of one kind of evidence. Low, 30%: some evidence, not enough to announce it. Never 100%.
- Effort. Trivial, under a developer-month (0.5); small, one to three (1); medium, a quarter or more (2); large, a year or more (4). The effort budget counts in these units.
How to score feature requests in Kjerr
The internal roadmap is a tab in the admin, next to the inbox. Nothing is set up first: the first roadmap exists when you open it.
- Add the ideas. Tick a roadmap on any idea in the inbox, or search the inbox from the roadmap's own add box. A new idea typed there lands on the board as well, so it can collect feature votes like any other.
- Pick the four values. Each cell is a drop-down with the scale's labels. The score and the sums above the table update as you go; a status change made here sends no email.
- Sort, group, filter. By score, votes, companies, MRR, owner or ETA; grouped by status, owner, ETA or topic; filtered to one owner, one status or the ideas above the line.
- Set the budget. Under Settings, an effort budget and the period it covers. The line appears, and Above the line only shows what fits.
- Share a view. Save the filters as a view and share it as a read-only link that needs no login, or export the table as CSV.
- Tell the voters. When a decision is made, change the status from the inbox, which emails the author and every voter, and write it up on the roadmap your customers see and the changelog.
A worked example
Two requests sit at the top of the inbox with thirty votes each. The first, exports to CSV, touches everyone (Reach 4), is a significant improvement (Impact 2), and the comments and the forty companies behind it leave little doubt (Confidence 80%); it is a quarter of work (Effort 2). Score: 4 × 2 × 0.8 ÷ 2 = 3.2.
The second, a native mobile app, would be a game changer (Impact 4) but for a segment (Reach 2), with thin evidence beyond the votes (Confidence 30%) and a year of work (Effort 4). Score: 2 × 4 × 0.3 ÷ 4 = 0.6. Same votes, five times the return on the work. With a budget of three points for the quarter, the line falls under the export and the mobile app waits, with its score and its reasons on the page for anyone who asks why.
Beyond RICE: your own factors, weights and a second opinion
RICE is the default, not the rule. The scoring page is a short form, and the AI only proposes.
- Add a factor such as strategic fit or risk, name its steps and say whether it raises or lowers the score. Rename or drop the defaults.
- Give a factor a weight. Under RICE a weight of 2 counts it twice as hard (its value is squared); 0 keeps the column but leaves it out of the score. Reset to RICE puts the defaults back.
- Suggest scores asks OpenAI to propose a value per factor for up to eight unscored ideas, each with a one-sentence reason naming the votes, companies and comments it leaned on. Proposals are ghosts until a teammate applies or dismisses them.
- Several roadmaps per workspace, one per team or per quarter, each with its own factors and budget. Archive one when the quarter ends.
What this RICE scoring tool leaves out
It ranks the requests on your board. It does not try to be a planning suite, and it never shows its working to customers.
- No scores, dates or estimates on the public roadmap. Customers see statuses.
- No sprint planning, assignees across tools or Jira sync. A webhook on status changes can feed the tracker you already use.
- No revenue model. MRR is shown beside each idea as the directory holds it; what it should weigh is your call, through Reach and Confidence.
- No scoring without a person. AI proposes; a teammate applies.
Who it is for
Product teams at SaaS companies of any size, from startups to enterprise, that want a public feedback board, a team-only roadmap scored with RICE, and reliable follow-up emails without a help-desk suite or a per-seat bill. RICE prioritization earns its place once the inbox holds more requests than the next quarter can take, and the question becomes which ones, and why. The feedback analysis tool says what the requests add up to; the roadmap says what to do first.
Questions, answered plainly.
What is RICE prioritization?
RICE prioritization is a way of ranking what to build next by four factors: Reach (how many people it touches), Impact (how much better things get for them), Confidence (how much evidence sits behind the first two) and Effort (how much work it is). The score is Reach × Impact × Confidence ÷ Effort, and a higher score means a better return on the work. It was written down at Intercom and is the most used prioritization framework in product teams.
How is a RICE score calculated in Kjerr?
Each factor is a pick from a short scale: Reach 4, 2 or 1; Impact 4, 2, 1 or 0.5; Confidence 80%, 50% or 30%; Effort 0.5, 1, 2 or 4. Kjerr multiplies Reach, Impact and Confidence and divides by Effort the moment a value changes, and hovering a score shows the arithmetic, for example 4 × 2 × 0.8 ÷ 2 = 3.2.
Is there a RICE calculator?
The internal roadmap is one. Every idea on it is a row with the four factors as drop-downs and the score beside them, recalculated as you pick. Sort by the score, and the table is the ranked list a RICE spreadsheet would give you, with the votes, companies and MRR behind each idea already filled in.
Do customers see the RICE scores?
No. The internal roadmap is visible to the team only: Viewers and above can read it, Moderators and above edit it. The public roadmap on the board shows statuses, never scores, dates or estimates. A saved view can be shared as a read-only link with people outside the team, for a planning meeting or an investor.
Can I use value/effort or my own factors instead of RICE?
Yes. Under Scoring you rename, add or remove factors, set each one's scale and give it a weight. A factor either raises the score or lowers it, so value ÷ effort is RICE with two factors, and strategic fit or a risk factor is one more row. Reset to RICE puts the defaults back.
Does the AI decide the priorities?
No. Suggest scores asks OpenAI, on servers with a key, to propose a value per factor for up to eight unscored ideas, with a one-sentence reason that names the evidence. The proposals sit as ghosts in the cells until a teammate applies or dismisses each one. Nothing is scored until a person says so.
Related pages
Public roadmap toolWhat customers see: statuses, never scores.
Feature request toolWhere the requests and their votes come from.
Feedback analysis toolInsights: what the feedback adds up to.
Customer feedback analysisFive steps, and what to count.
Give your users an answer.
Kjerr is free during early access, with every feature included and no credit card. When paid plans arrive they will be priced per workspace, never per admin seat and never per tracked user, so the number of people who post, vote or comment never changes the bill.