• RICE: SIMPLE priorities FOR PRODUCT MANAGERS
  • The Nuggets translation Project
  • Translator: ZhaofengWu
  • Proofreader: Lynulzy, Dongpeiguo

Priorities are always a challenge when establishing a product roadmap. How do you decide where to start?

If you’ve put in enough effort to come up with new ideas, look for opportunities for improvement, and gather feedback, you’ll have a solid road map of good ideas. But the order in which you approach these ideas is also worth thinking about. You need to take some time to properly prioritize.

Priorities are a conundrum

Why is it so hard to sequence the product roadmap? Let’s list them:

  • It’s much more satisfying to implement your own small, personal idea than a broader project.
  • It’s tempting to focus on smart ideas rather than projects that directly impact your goals.
  • Jumping into new ideas is more exciting than working on projects that are already in the works.
  • It’s easy to overlook that some projects may require more effort than others.

Even if you can overcome these problems perfectly, you will always need to combine and compare these factors in all your project ideas. The good news is, you don’t have to do this in your head.

A simple prioritization tool

That’s where a scoring system comes in. A good scoring system helps you clearly and systematically consider each of these factors in a project idea, and integrate them tightly and uniformly.

Using such a scoring system to establish priorities is not new. There are now many systems to balance the books. However, my team at Intercom did not find a system that would allow us to compare ideas uniformly.

So last August, starting from the ground up, we set out to develop our own priority scoring system. After many tests and iterations, we identified four factors and a way to combine them.

RICE: Four factors to evaluate priorities

RICE is an acronym for the four factors we use to judge project ideas: reach, Impact, confidence, and effort.

The scope of

To avoid bias based on personal factors, you should first estimate the scope of each project’s impact over time. For example, for our team, we need to estimate how many users our project will affect in a quarter.

Ranges measure the number of people or events in each time period. It can be “customers per quarter” or “transactions per month.” Try to use realistic product data rather than random numbers.

The sample

Project 1:500 customers arrive at this registration process per month, of which 30% choose this option. So the range is 500 × 30% × 3 = 450 customers per quarter.

Project 2: Every quarter all customers using this feature will notice this change. So the range is 2,000 customers per quarter.

Project 3: This change will only have a one-time impact on our current 800 customers and will not last. So the range is 800 customers per quarter.

Impact strength

In order for the project to accurately achieve your goals, you need to consider the impact of the project on the individual. The question our team has to ask is, how much does it increase conversion rates when customers use our program? Of course, your team should change this to your goal, such as improving usage or mood.

The strength of the impact is hard to measure precisely. So, I took the form of a scale: 3 for “great impact,” 2 for “high,” 1 for “medium,” 0.5 for “low,” and 0.25 for “almost none.” When calculating the final score, the numbers are multiplied by the coefficients to zoom in and out.

It may seem unscientific to choose a number for the strength of the impact, but otherwise using intuition can be messy.

The sample

Project 1: It has a huge impact on every customer. The impact score is 3.

Project 2: There will be a small impact on each customer. The impact score is 1.

Project 3: The impact is somewhere in between. The impact score is 2.

confidence

In order to control your enthusiasm for exciting but vague ideas, you should consider your faith in some estimates. If you think a project is going to have a huge impact, but don’t have any data to back it up, this confidence factor can help you control your impulses.

Confidence is a percentage, and I also use a scale to avoid the difficulty of choosing. 100% is “high confidence,” 80% is “medium,” and 50% is “low.” Lower than that? That’s nonsense. But be honest with yourself: How much is behind your estimate?

The sample

Project 1: We had quantitative calculations of scope, user surveys of impact strength, and engineering assessments of effort effort. This item can have a 100% confidence score.

Project 2: I have data on the support range and energy, but I’m not sure about the impact. This project has an 80 percent confidence score.

Item 3: Scope and impact might be a little lower than expected, but energy might be a little higher. The project can only have a 50 percent confidence score.

energy

To make an impact with minimal effort and the fastest pace, you need to estimate how long a project will take. You need to estimate from the perspective of each team member: product person, designer, and engineer.

Energy is measured in “man months” — the amount of work a team member can do in a month. There are a lot of unknowns, so I estimate the amount roughly as an integer (or 0.5 for much less than a month’s work). Unlike other positive factors, more energy is a bad thing. So, we need to divide the total influence by this number.

The sample

Project 1: This project takes about a week of planning, a week or two of design, and two to four weeks of engineering. I think the two-month score is more appropriate.

Project 2: This project takes several weeks to plan, a longer period to design, and at least two months for each engineer. This project is about 4 person months of energy score.

Project 3: This project required only one week of planning, no new design, and only a few weeks of engineering time. So it’s just a fraction of a month.

The RICE score was obtained by combining the factors

To summarize our four factors:

Scope: How many people will it affect? (within a fixed time frame)

Strength of Influence: How much does it affect a person? (Large = 3x, large = 2x, medium = 1x, low = 0.5x, very low = 0.25x)

Confidence: How confident are you in your estimates? (High = 100%, medium = 80%, low = 50%)

Energy: How many man-months will it take? (Use whole numbers, or in very small cases, 0.5 — not too precise.)

Once you’ve estimated these factors, combine them into a single score so you can compare multiple items at once. The formula is as follows:

The resulting score measures “impact per unit of hours worked” — which is exactly what we want to maximize. I made a spreadsheet to automatically calculate the score along with my estimates.


You can make a copy of this form for your own use. You can alsoDownload a version of.xls

Once the score has been calculated in the first step, you can sort your list and reevaluate. Were there any events that were overscored or underscored? If so, reconsider your estimate and either change it or accept that your intuition may not be right.

I have used this RICE score to evaluate ideas for more than 100 projects over the past six months. It has helped me enormously in deciding on ideas that are hard to compare. It forces you to think about why an idea for a project has impact, and it makes you realistic about the effort involved.

Use RICE fractions effectively

Of course, RICE scores should not be regarded as a golden rule. Some of the low-scoring items may also have good reasons to do them first. For example, one project may depend on another, so the latter needs to be done first. Or, a feature might be a selling point to a target customer.

Our team sometimes runs out of RICE’s score order, which is perfectly fine. The scoring system just gives us a clearer view of the trade-offs.

A scoring system like RICE’s will help you make more informed decisions and explain to others why you are doing so. In your own prioritization process, try RICE and let us know how it works!