RICE Prioritisation: How to Score Ideas You Can Defend

RICE is a simple way to score and rank product ideas using four factors: reach, impact, confidence and effort. You multiply the first three together and divide by effort. The highest score goes to the top of the list.

It won't make the decision for you. What it does is turn a debate about opinions into a conversation about numbers everyone can see and question. That alone is worth a great deal.

What RICE Means

The framework was first described by the product team at Intercom. They wanted a quick, repeatable way to compare ideas that looked nothing alike. Each letter is one input.

• Reach. How many people will this touch in a set period? Use a real count where you can, such as customers per quarter or sessions per month.
• Impact. How much will it help each person it reaches? Use a simple scale: 3 for massive, 2 for high, 1 for medium, 0.5 for low and 0.25 for minimal.
• Confidence. How sure are you about the reach and impact figures? Use a percentage: 100% for strong evidence, 80% for decent evidence and 50% for a hunch.
• Effort. How much work will it take across everyone involved? Use person-months; rough is fine.

The score is reach × impact × confidence ÷ effort. That's all there is to the arithmetic.

When to Choose RICE

Choose RICE when you need a quick, repeatable way to score and compare a list of ideas. It works best when you have a decent number of ideas, some data about your users and a team that keeps arguing in circles.

It's especially good at one thing. It lets small changes that reach everyone compete fairly with big, exciting features that reach a few. Without a reach column, the shiny feature usually wins the meeting.

It's less useful when timing is the main issue. If one idea becomes worthless after a fixed date, RICE won't show it. That's a job for Cost of Delay.

How to Use RICE, Step by Step

1. Agree the time period for reach. Pick one, such as a quarter, and use it for every idea.
2. Estimate reach for each idea. Use your analytics where you can. Write down the source of each number.
3. Rate impact on the shared scale. Talk about what "high" means for your product before anyone scores.
4. Set a confidence level. Be honest. A low number isn't a failure. It's useful information.
5. Ask the people who'll build it for effort. Person-months, total, not per discipline.
6. Calculate and rank. Then look for surprises. Those are where the best conversations happen.
7. Revisit as new evidence arrives. Scores should change when you learn something.

A Worked Example

Imagine an online garden centre with about 40,000 customers a quarter. The team has four ideas on the table.

Idea	Reach	Impact	Confidence	Effort	Score
Show delivery dates on product pages	40,000	1	80%	1	32,000
Plant care reminders by email	12,000	2	50%	2	6,000
Augmented reality garden planner	3,000	3	50%	6	750
Faster checkout on mobile	25,000	1	80%	3	6,667

The garden planner was the idea everyone loved. It scored lowest by a long way. Delivery dates on product pages, a modest change, came out on top because every customer sees it.

Notice the confidence column too. The care reminders scored well, but at 50% confidence. The right next step isn't to build them. It's to run a small test that would raise or lower that number.

Running a RICE Session with Your Team

RICE works best as a short, regular habit. Here's a simple way to run it in under an hour.

Before the session, the product manager pulls reach figures from analytics for each idea. Engineers give rough effort estimates. That way the meeting is spent on judgement, not on hunting for data.

In the session, score impact first. Ask each person to write their rating down privately, then reveal together. Where people disagree, ask the highest and lowest scorers to explain. Their reasons are usually more useful than the final number.

Then agree confidence. This is the column that keeps everyone honest. A good prompt is simple: what evidence do we actually have? If the answer is "a conversation with one customer", 50% is generous.

Finally, calculate and look at the ranked list together. Ask two questions. Does anything near the top surprise us? Does anything near the bottom feel wrong? If an idea feels wrong, look at its inputs. One of them is usually a guess dressed up as a fact.

Keep the scored list somewhere visible and update it as new evidence arrives. A RICE list that's months old tells you very little.

Common Mistakes

• Inflating confidence. It's tempting to nudge it up to help a favourite idea. If confidence is low, the next step is a test, not a build.
• Treating the score as the decision. RICE starts a conversation. Strategy, dependencies and timing still matter.
• Mixing reach periods. One idea scored per month and another per year will give nonsense rankings.
• Letting one person score everything. Ask the people closest to the data and the build.
• Hiding the inputs. Always show the four numbers next to the score, so people can challenge the right one.

How RICE Compares with Other Prioritisation Tools

RICE and WSJF both divide value by effort. WSJF adds time urgency and risk reduction, so it's better when timing matters. RICE is simpler and quicker for a list of product ideas.

RICE and MoSCoW do different jobs. MoSCoW sorts what goes into a fixed release. RICE ranks a list of ideas against each other.

RICE and Cost of Delay also differ. Cost of Delay puts a money value on waiting. If you can estimate that value, CD3 often gives a sharper order than RICE.

In my experience, teams get the most from RICE when they use it every week for a few minutes, not once a quarter in a long workshop. The habit matters more than the precision.

If you'd like to try it on your own list, the Product Field Kit includes a fillable RICE worksheet and a calculator that ranks your ideas as you type.

Questions People Ask

← More Field Notes