No team has time to do everything it wants. The list of ideas, requests and improvements is always longer than the available capacity — and that is precisely why the hard question is rarely "what can we do?", but rather "what do we do first?".
Without a method, that decision tends to fall the wrong way: you do what the loudest voice asked for, what arrived last, or whatever the boss mentioned in the corridor. A prioritisation matrix does not remove human judgement, but it forces you to make it explicit. Instead of arguing opinions, the team compares initiatives against agreed criteria — and the choices become defensible, repeatable and easy to communicate.
This guide runs through the most useful matrices — Impact × Effort, RICE, MoSCoW and weighted scoring — explains when each makes sense, shows how to run the exercise with the team and points out the mistakes that hollow out the method. At the end, a generic case shows how honest prioritisation changed the course of a quarter.
Why you need a prioritisation matrix
The problem with prioritising "off the top of your head" is not a lack of intelligence — it is an excess of biases. We give too much weight to the most recent request, to the customer who called today, to the idea from whoever holds most power in the room. This is called deciding by the HiPPO (the highest-paid person's opinion) or by whatever is freshest in memory. The result is a roadmap that reflects who has influence, not what generates the most value.
A matrix fixes this in three ways: it makes the criteria explicit (what counts as "important"?), it forces you to compare everything on the same scale, and it leaves a trail — three months from now you can see why that item ended up at the top. It is not about maths; it is about making the conversation honest.
Impact × Effort: the simplest matrix
If you only use one, use this. You draw a square with impact on the vertical axis and effort on the horizontal, and place initiatives into the four quadrants:
- High impact, low effort — the "quick wins". Do these first, no debate.
- High impact, high effort — the "big bets". Worth it, but they demand planning and commitment.
- Low impact, low effort — the "fill-ins". Do them when there is spare time.
- Low impact, high effort — the "money pit". Avoid them, however appealing they look.
The strength of this matrix is speed: in half an hour, a team can place twenty initiatives on the map and immediately see where to start. The weakness is the subjectivity of "impact" and "effort" — which leads us to the next matrices.
RICE: when you want more rigour
RICE breaks the vague "impact" down into four more concrete factors: Reach (how many people or cases are affected in a period), Impact (how much changes for each one), Confidence (how sure we are of the estimates) and Effort (the cost, usually in person-months). The score combines them:
RICE = (Reach × Impact × Confidence) ÷ Effort
The value of RICE lies less in the final number and more in the discipline of estimating each factor separately. Confidence, in particular, is a precious brake against enthusiasm: an idea with enormous theoretical impact but low confidence in the estimates naturally drops down the list. Use RICE when you have many comparable initiatives and want a finer ranking than the 2×2 allows.
MoSCoW: prioritising by obligation, not by score
Not everything lends itself to scoring. When what is at stake is the scope of a delivery — what goes in and what stays out of a release — MoSCoW usually serves better. It classifies each item into four levels: Must (without this, the delivery fails), Should (important but not blocking), Could (desirable if there is room) and Won't (explicitly left for later).
The discipline of MoSCoW is in the "Won't": naming what you are not going to do is as important as choosing what you do, and it prevents the silent creep of scope. A healthy rule is to limit the "Must" items to a fraction of total effort, to leave slack for the unexpected.
Weighted scoring: when the criteria are several
Sometimes "impact" is not enough, because there are several objectives in play — revenue, risk, customer satisfaction, strategic alignment. The weighted scoring model handles this: you choose the criteria, assign a weight to each (adding up to 100%), score each initiative on each criterion and calculate the weighted average.
The trick is not in the formula, it is in the choice of weights — that is where strategy enters the calculation. If "risk" weighs 40%, you are telling the team that this quarter is for consolidating, not for taking risks. Setting the weights before seeing the initiatives avoids the temptation to adjust them to justify the answer you already wanted.
How to run the exercise, step by step
The matrix is only the tool; the value is in how you use it. A process that works:
- Define the objective for the period first of all — without a clear goal, "impact" means nothing.
- List the initiatives at the same level of granularity (do not mix "change a button" with "enter a new market").
- Choose a single framework suited to the decision and fix the scales before scoring.
- Score together, or score separately and compare — the disagreements are the most informative part of the meeting.
- Decide and record why, not just the final order.
- Review regularly: priorities change when new information arrives.
Common mistakes to avoid
The most frequent is false precision: three decimal places on a score built from guesses. The numbers are there to rank and to reveal disagreements, not to fake objectivity. Next comes scoring in isolation — when a single person scores everything, the matrix merely dresses up their bias as rigour. Then there is treating prioritisation as a one-off event, when it should be a habit. And finally, ignoring effort and confidence: ranking by impact alone fills the top of the list with enormous, uncertain projects that never leave the drawing board.
Mini case: the quarter that changed course
A B2B software company started each quarter with a list of features requested by customers and prioritised it, in practice, by the noisiest requests — the ones that came up in renewal calls. The roadmap was always chasing fires, and the higher-reach initiatives were left for "next quarter", which never came.
The team switched to RICE. When estimating Reach, it realised that three of the most insistent requests came from two large customers but affected less than 5% of the base. In contrast, an improvement to the onboarding process, which nobody asked for out loud, touched practically every new customer and took moderate effort. The RICE score put it at the top, well above the noisy requests.
The result, one quarter later: the average time to first value for new customers fell by about 30%, and the activation rate rose visibly. No decision was "against" the large customers — it simply became explicit that the reach of the onboarding improvement was far greater. The matrix did not discover anything the team could not have known; it forced them to look at the numbers before looking at who spoke loudest.
In practice
A prioritisation matrix is worth the conversation it forces, not the score it produces. Start with Impact × Effort to gain speed; move to RICE when you need to rank many initiatives with more rigour; use MoSCoW to define the scope of a delivery; and turn to weighted scoring when there are several objectives to balance. Whatever the choice, fix the criteria before looking at the options, score as a team and record the why. The goal is not the perfect matrix — it is to stop deciding by the most recent or the loudest and start deciding by what generates the most value.
