How to Run a 5-Day Innovation Sprint in a Small UK Lab

A five-day sprint in a lab the size of a large cupboard sounds like a compromise. It isn't. The constraint does a lot of the work for you: fewer people, less kit, one question, and a deadline everyone can see from their own desk. Here's the week we run when something needs an answer quickly and there's no budget for a three-month programme.

Start with the question, not the workshop

The most common way a sprint fails is booking the week before anyone knows what they're testing. Spend an hour on the Thursday before. Write the question as one sentence a stranger could understand: "Can a first-time visitor find the recycling point without asking?" Not "explore the recycling experience". If the answer already exists in a report, a support inbox or a colleague's head, don't spend a week on it. Go and read the report.

A test that fits in five days

Before you commit, check the idea against three filters:

• You can build something rough in two days or less.
• Someone outside the team can react to it by Thursday.
• The result could change a decision you are already facing.

If it fails any of those, shrink it until it passes. A cardboard mock-up and four conversations beat a polished prototype and nobody to show it to.

Plan the week as an arc

Days in a small lab fill up with ordinary work if you let them. Give each day one job.

1. Monday — frame and cut. Agree the question, decide what you'll show and to whom, and write "we'll know we were wrong if…" on the wall. Book your Thursday participants before lunch.
2. Tuesday — build rough. Get something in front of a colleague by the end of the day, however ugly.
3. Wednesday — test internally and fix. Run it past two people who weren't involved. Watch where they hesitate, then make the smallest change that removes the confusion.
4. Thursday — go external. Run your sessions one at a time and write raw notes the same afternoon, while the details are still sharp.
5. Friday — review and decide. Half a day, no building. This is the part teams skip and then regret.

Make one room do three jobs

Space is usually the tightest constraint. You don't need a studio; you need zones that don't fight each other. A bench for making, a corner with two chairs and no screens for conversations, and one wall that becomes the show wall — the only place work is allowed to live for the week.

Three habits keep a small room usable:

• Clear the benches at the end of every day, even mid-build. Ten minutes, everyone helps.
• Label crates and stack them vertically; horizontal piles eat floor space.
• Have a sound plan. Two pairs of ear defenders and a phone playing brown noise will do more for concentration than any policy.

The room should look different at 9am and at 5pm. If it doesn't, nothing happened that day.

Facilitation when you're also building

In a lab of four people, nobody can spend the week purely facilitating. Rotate the role daily and keep it light. Whoever holds it does three things: watches the clock, writes decisions into one shared log, and asks the awkward question when the group drifts.

Keep a parking-lot sheet on the wall for ideas that are good but out of scope; it stops the same tangent arriving three times. Watch for the two failure modes small teams fall into — the loudest person designing by volume, and quiet agreement that isn't agreement. Ask directly: "What would make us stop?" and "Who disagrees?" Then wait. Silence after a question usually means someone is thinking, not that they've finished.

Kit, cost and what to borrow

You almost certainly have enough already. A working printer, flipchart paper, marker pens, tape, sticky notes, a timer and a phone that records audio will carry the week. Borrow the rest from another team and return it on Friday.

Spend money on two things only: whatever the test genuinely requires — a specific material, a small piece of hardware, a printing run — and making participants comfortable. Buy the good biscuits. If you're paying incentives or handling anything personal, check your organisation's policy first and follow your data protection obligations; if you're unsure, ask the relevant team before Thursday, not after.

Friday: what counts as a result

A result is an observation, not a feeling. "Three of five people ignored the second screen" is a result. "The session felt positive" is a mood. Aim for at least one observation that would have been uncomfortable to predict on Monday.

Write a one-page summary before anyone leaves the room:

• The question, in its original words.
• What we built and who we showed it to.
• What we saw, including what didn't work.
• What we now believe, and how confident we are.
• What happens next, with a name and a date.

Negative results deserve the same treatment as positive ones. A sprint that kills a weak idea in five days has done its job; a sprint that produces a polite maybe has not.

The Monday that decides whether it mattered

Book half an hour for the following Monday and treat it as part of the sprint, not an optional extra. Walk through the summary, confirm the owner of each next step, and put the next test in the diary. Then close the loop: archive the notes and photos somewhere searchable, reset the room, and note the two things you'd run differently.

Sprints are tiring in a way that spreadsheets are not. Don't schedule two back to back, and keep the Monday afterwards light. If the week worked, the reward isn't a longer sprint — it's a shorter one next time, with a sharper question and fewer people in the room.

Photo: ClickerHappy / Pixabay