Improvement Planner
Add the usability problems you know about, rate each one for impact, effort and evidence, and get a plan that sorts them into six groups, from the changes to make first to the ones to park.
Short answer
For each problem, describe it, note where it happens and rate impact, effort and evidence. Select Build the plan to sort the list into six groups. Items that rest only on opinion are flagged so you can check them before building anything.
On this page
Exporting needs JavaScript, and so does automatic sorting. You can apply the same rules by hand: rate each issue for impact and effort, then use the grid described on this page to put it in one of six groups.
How the planner sorts your list
The planner places each problem in one of six groups based on two ratings: its impact on visitors and the effort needed to fix it.
| Impact | Low effort | Medium effort | High effort |
|---|---|---|---|
| High | Do first | Do first | Plan as a project |
| Medium | Quick improvements | Schedule later | Schedule later |
| Low | Fix when you are nearby | Park for now | Park for now |
High-impact problems that are reasonably easy to fix come first. High-impact problems that need major work become projects with their own scope, because squeezing them into a sprint risks a half-finished fix. Low-impact items are not ignored; they are grouped so your team can fix them while working on related pages.
Rating impact
Impact describes how much a problem hurts visitors trying to complete a task. Use High when the problem blocks the task or sends people the wrong way, Medium when it slows people down or causes doubt but they recover, and Low for minor annoyances. Think about how many visitors meet the problem as well as how badly it affects them.
Rating effort
Effort describes the work needed to fix the problem. Low covers copy changes, settings and small styling fixes. Medium covers changes that need design and development time. High covers rebuilds, new features or changes that depend on another team or vendor. Ask the people who will do the work; their estimates are more reliable than anyone else's.
Why evidence matters
The evidence rating does not change which group an item lands in, but it changes how much you should trust that placement. A problem seen in several usability sessions is far more certain than one based on a colleague's opinion. When an item rests on opinion or assumption, the plan marks it "verify first" and adds a note at the end. A quick check, such as watching three people try the task or looking at the relevant analytics, can save a week of development on a problem that is not real.
Using the plan
Export the plan as a text file or copy it into your project tracker. Review it with your team, adjust any ratings you disagree on and assign owners for the Do first group. After the changes ship, check whether they helped, then update the ratings. Our recommendations service produces a fuller version of this plan, with acceptance criteria and a way to check each change.
Privacy and accessibility
The planner runs entirely in your browser and stores nothing. All fields and buttons work with a keyboard, errors are announced and linked to the field, and the list updates with a status message after each change.
Using the planner
Can I edit an item after adding it?
Remove it and add it again with the new ratings. Keeping the tool simple means there is no editing mode.
What if everything is high impact?
That is a sign that impact may be rated by how much the team cares, not how much visitors are affected. Compare items against each other and keep High for problems that block tasks.
Is this the same as an ICE or RICE score?
It is simpler. Scoring formulas add numbers that can suggest false precision. The grid keeps the conversation about two questions the whole team can discuss: how much does it hurt, and how hard is it to fix.