Every service design project reaches the same moment. You’ve done the research, you’ve mapped the journey, and the wall is covered in sticky notes full of good ideas. Then someone asks the question nobody wants to answer: ‘We can’t do all of this, so what are we actually going to do first?’
That’s where MoSCoW comes in. It’s one of the simplest prioritisation tools there is, and in my experience one of the most useful. I’ve used it with charities, community groups and public services to turn a long list of wants into a plan that people can actually deliver, and, just as importantly, to have honest conversations about what isn’t going to happen yet.
In this article, I’ll explain what MoSCoW prioritisation is, how I run it in service design workshops, and the lessons I’ve learned the hard way.
What MoSCoW stands for
MoSCoW sorts everything you could do into four groups. The capital letters give the method its name; the small o’s are just there to make it pronounceable.
| Category | What it means | A useful test |
|---|---|---|
| Must have | The service won’t work, or won’t be legal or safe, without it | If this were missing on launch day, would we delay the launch? |
| Should have | Important and painful to leave out, but the service still works | Is there a workaround, even an awkward one? |
| Could have | Wanted or desirable, but with less impact if it’s left out | Would most users notice if it never arrived? |
| Won’t have (this time) | Agreed not to happen in this phase, but not forgotten | Have we written down why, and when we will look at it again? |
MoSCoW was created by Dai Clegg at Oracle in the 1990s and became part of DSDM, an agile project framework now looked after by the Agile Business Consortium. In DSDM, the Must Haves are described as the Minimum Usable SubseT: the smallest set of things that still gives people a usable service. The Consortium suggests keeping Must Haves to no more than about 60% of the effort, so that the Shoulds and Coulds act as contingency if things take longer than planned.1
Why MoSCoW works so well in service design
Service design brings together lots of people with different priorities: the people who use the service, frontline staff, managers, IT, finance and sometimes funders or councillors. Everyone arrives with a list. MoSCoW gives them a shared language for deciding between those lists.
- It’s easy to explain: I can introduce MoSCoW in two minutes to a room of people who have never heard of it, including service users who’ve never been to a workshop before.
- It forces real trade-offs: a long list of ‘high priority’ items helps nobody. Four clear categories make people choose.
- It makes saying no feel safe: ‘Won’t have this time’ is much easier to agree to than ‘no’. People are more willing to let go of an idea when they know it’s been recorded and will be revisited.
- It keeps users at the centre: if you sort user needs rather than features, the conversation stays focused on what people need from the service, not what the organisation would like to build.2
How I run a MoSCoW session
This is the approach I’ve settled on after running MoSCoW sessions with all kinds of groups. Allow around two hours for a typical service, and more if the list is long or the stakes are high.
Step 1: Start with evidence, not opinions
Before the session, gather what you’ve learned from user research: interviews, journey maps, feedback, complaints and anything from activities such as Photovoice. If you prioritise before you understand people’s needs, you’re really just ranking opinions, and the loudest voice usually wins.
Step 2: Write each item as a need
Put each requirement on its own card or sticky note, written as a need rather than a solution where you can. ‘People can book a room without phoning during office hours’ leaves room for good ideas. ‘Build an app’ doesn’t. Keep the wording short and plain, so everyone in the room understands every card.
Step 3: Agree what each category means
This is the step people skip, and it’s the one that matters most. Agree the timeframe you’re prioritising for, such as the first release or the next six months, and agree what makes something a Must. I write the tests up on a flipchart: ‘Would we delay the launch without it?’, ‘Is there a workaround?’ and ‘Is it a legal requirement?’. If there’s a workaround, it isn’t a Must.
Step 4: Sort in silence first
I ask people to place the cards in the four columns on their own, without talking, for ten minutes or so. Sorting in silence stops the most senior person in the room setting the tone, and quickly shows where people agree. Cards that everyone puts in the same column can usually stay there.
Step 5: Discuss the cards that moved
Spend your discussion time on the cards that people placed in different columns, or that kept being moved. Ask the person who placed it to explain why, and then use the tests from Step 3. If you’re stuck, dot voting or an Impact/Effort Matrix can help the group decide.
Step 6: Challenge the Musts
Almost every group puts too much in the Must column on the first pass. Go through each Must one at a time and ask, kindly but firmly, ‘What would actually happen if we launched without this?’. Keep going until the Musts are something the team is confident it can deliver, with room to spare.
Step 7: Record the Won’ts and the reasons
Write down every Won’t have, why the group made that choice and when you’ll look at it again. This is what makes the Won’t column trustworthy. I often keep a Parking Lot on the wall for ideas that come up during the session, so nothing gets lost.
Step 8: Revisit it
MoSCoW isn’t a one-off. Priorities change as you test with users and learn more, so revisit the list at the start of each new phase. A Should can become a Must when testing shows people can’t manage without it, and a Must sometimes turns out to be less important than everyone thought.
A worked example: booking a room at a community centre
To show how this works in practice, here’s a simplified example of the kind of project I often work on: a community centre redesigning how local groups book its rooms. At the start, bookings were made by phone during office hours, recorded in a paper diary, and double bookings were common.
| Category | Examples |
|---|---|
| Must have | See which rooms are free and request a booking without phoning; booking confirmation by email or text; no double bookings; works with screen readers and on a phone; a way to book for people who aren’t online |
| Should have | Pay online; cancel or change a booking yourself; regular weekly bookings set up in one go |
| Could have | Photos and floor plans of each room; reminders the day before; a waiting list for popular slots |
| Won’t have (this time) | An app; integration with the council booking system; discounts calculated automatically for charities |
Notice a few things. Accessibility and an offline route are Musts, not nice-to-haves: if some people can’t use the service, it isn’t working. Online payment is a Should because the centre could keep taking payment the old way for a while. And the app is a Won’t this time, not because it’s a bad idea, but because a mobile-friendly website meets the same need for far less effort.
Lessons I’ve learned
Everything becomes a Must
This is the most common problem by far. When everything is a Must, nothing is, and the plan quietly becomes impossible. The tests in Step 3 are your best defence, along with the 60% guideline. I’ve found it helps to ask people to imagine they have half the budget they expected: the real Musts become obvious very quickly.
The highest-paid person in the room
Prioritisation brings power into the room. If a director says something is a Must, it’s hard for a frontline worker or a service user to disagree. Silent sorting helps, as does making sure people who use the service are in the room, or that their words are, through quotes and research findings on the wall.
Won’t have doesn’t mean never
If the Won’t column becomes a place where ideas go to die, people stop trusting it and start fighting to get everything into the Musts. Always go back to it at the next review, and tell people what happened to their ideas.
Some things aren’t up for debate
Legal requirements, safety and accessibility belong in the Must column. In the UK, the Equality Act 2010 applies to services, and public sector websites and apps also have to meet accessibility regulations.3 It’s worth saying this at the start, so nobody is tempted to trade accessibility away for a shiny feature.
Sort needs, not solutions
When people sort solutions, they argue about technology. When they sort needs, they talk about people. If a card names a specific product or feature, ask ‘What need is this meeting?’ and rewrite it before it’s sorted.
Tips for facilitating MoSCoW
- Use colour: give each category its own colour of card or column header, so the shape of the plan is visible from across the room.
- Keep the cards small: one need per card, so each one can move on its own.
- Bring in the people who use the service: their view of what’s a Must is often very different from the organisation’s, and it’s usually more accurate.
- Timebox the discussion: give each disputed card a few minutes. If it still divides the room, note the disagreement and decide how it will be settled, rather than letting one card take over the session.
- Photograph the wall: and type it up the same day, with the reasons for each Won’t, so the decisions don’t get reopened by people who weren’t there.
MoSCoW won’t make difficult decisions for you, but it gives everyone a fair, clear way to make them together. Used well, it turns a wall of sticky notes into a service that can actually be delivered, and leaves people feeling heard even when their idea didn’t make the first cut.
What does MoSCoW stand for?
MoSCoW stands for Must have, Should have, Could have and Won’t have (this time). The two o’s are added to make the word easy to say. It’s a way of sorting requirements so a team can agree what to deliver first.
How many Must haves should there be?
Fewer than most groups think. The Agile Business Consortium, which looks after the DSDM framework where MoSCoW comes from, suggests keeping Must Haves to no more than about 60% of the effort, leaving the rest as contingency.
How do you decide if something is a Must have?
Ask what would happen if it were missing. If you would delay or cancel the launch without it, or it’s a legal or safety requirement, it’s a Must. If there’s a workaround, even an awkward one, it’s a Should or a Could.
Can you use MoSCoW with service users?
Yes, and it works well. The four categories are quick to explain in plain English, and including the people who use a service often changes what the group sees as a Must.
Sources
- Agile Business Consortium: What is MoSCoW Prioritization? The definitions, the tests for a Must Have and the guidance on effort. ↩︎
- GOV.UK: Service Standard. The standard UK government services are assessed against, starting with understanding users and their needs. ↩︎
- GOV.UK: Accessibility requirements for public sector websites and apps. ↩︎



