What to change
Show the underlying counts beside the percentage and review whether the observation window is sufficient for the decision. A small numerical difference can produce a dramatic relative change without establishing a stable pattern. Ask the analyst to compare a longer relevant window and document operational changes that might explain the result. Avoid inventing a statistical certainty; the useful output is a proportionate next check based on the available evidence.
A worked example
Illustrative example
Illustrative test: A fictional change from two to three observations is described with both counts and its percentage.
How to check the result
The manager understands the scale and limits before commissioning a fix.
A mistake to avoid
A large percentage is not automatically a large operational change.
Your action checklist
Use these checks to prepare a discussion with your team or supplier. Tick a check when you have recorded the decision or evidence, rather than when a feature has simply been promised.
Entries stay in this page and are not submitted to Kay & Co. Download your notes before leaving; the page does not save them.
Further reading
These primary sources provide additional context for the project decisions above.
Related decisions
Turn the brief into a working service
Kay & Co. can help you scope the work, design the experience and deliver the right solution for your organisation.
Explore marketing & search services or discuss your project.
Try the free Healthcare Digital Planner to find your starting priority.