FreeNo sign-upYour scores stay on your computer
ICE Score Calculator
Rank a shortlist of ideas by Impact, Confidence and Ease, each person scoring alone. It averages the scores, multiplies them, and flags where people's Confidence is far apart, so you test before you choose.
This tool also runs as a stage in a playbook, which takes you through several tools in order:
- Develop a new product idea Stage 5 of 6: which idea goes forward? Start at stage 1
Not following the playbook?
The outcome
- The one result the ideas should move. Impact is scored against it, so name it first.
Pick the shortlist from the earlier stages
- Four to six ideas that answer the problem. Ticking one adds it to the shortlist below.
The shortlist
Who scores
- Each person scores alone, then you compare. A role or initials is enough.
- Where people's Confidence scores are far apart, they are disagreeing about evidence: the tool shows those.
Your scores never leave your browser
The ideas, the people and the scores stay on your computer and in your progress file. To write the diagnosis, only the numbers are sent to our server. How it works
Each person scores alone, 1 to 10
Scored against: . Then multiply: the tool averages each factor across people.
- Impact
- If it works as hoped, how much does it move the outcome?
- Confidence
- What evidence says it will work?
- Ease
- How little work is it? 10 is very little.
How to score Confidence
1 to 2: an opinion. 3 to 4: what others say, or a similar product elsewhere. 5 to 6: data from your own customers. 7 to 8: a test with real customers. 9 to 10: people have paid for it.
Where Confidence scores are far apart
The ranking
Each factor is averaged across the people who scored; the ICE score multiplies the three. The highlighted row goes forward.
What goes forward
The checks
Save your progress
Download your progress file to continue from this point any time, on any computer: choose it under "Coming back?" at the start of the tool. Each new file includes everything so far, including the earlier stages when they are connected, so keep the latest one. Your work stays in this browser only until you close it.
Not saved yet.
Take your results with you
A PDF of the ranking, what goes forward and what it rests on, the checks, and the diagnosis.
About the tool
Why it exists
A score that shows where the evidence is thin
ICE ranks ideas by three factors anyone can score: Impact, Confidence and Ease. Scored together in a meeting, it becomes a ritual: a precise number on a guess, set by whoever speaks first. Scored by each person alone and compared, it shows something more useful: where people disagree about the evidence. This tool collects each person's scores, averages them, and stops you ranking while the Confidence scores are far apart, until a cheap test settles them.
It works on its own, and as stage 5 of our playbook for developing a new product idea, where the shortlist is picked from the brainstorm and SCAMPER, and the idea that goes forward goes on to a Business Model Canvas.
The method
How the tool works
The outcome first. Impact means impact on something, so the one outcome the ideas should move is named before anything is scored.
The shortlist. Four to six ideas, up to eight. In the playbook you tick them from stage 4's ideas (by prompt) and stage 3's (by theme); you can also type them.
Scoring alone. Each person scores every idea from 1 to 10: Impact (if it works as hoped, how much does it move the outcome?), Confidence (what evidence says it will work?) and Ease (how little work is it?). Each factor is averaged across the people who scored, and the ICE score is the three averages multiplied.
Where people disagree. When one idea's Confidence scores are 4 or more points apart, the people scoring are disagreeing about evidence. If that idea is in the top three, or would be with the highest Confidence anyone gave it, the tool asks for a cheap test first: what you did, what happened, and the Confidence the team now agrees, which then replaces the average.
The checks. The outcome named first, scored alone, disagreements tested, a clear lead (the top two more than a tenth apart), and a shortlist of four to six.
What leaves your computer. The ideas and names stay in your browser and your progress file. To write the diagnosis, the page sends our server only the numbers, by place in your shortlist, and the server keeps nothing.