Set realistic tasks
Translate specialist activities into a prototype test.
An operational planning prototype was tested on forecast review, limits and status changes before further design refinement.
Method, sample and scope make the evidence meaningful. Here is how the work was structured.
Moderated, task-based usability testing
Moderated sessions using a user-testing tool; physical setting not specified
Forecast review, planning navigation, limit setting and status changes
Task completion, reported ease-of-use ratings and observed failures
Translate specialist activities into a prototype test.
Record completion and ease ratings while watching the interaction.
Connect missed controls and misplaced expectations to design changes.
| Measure | Completion |
|---|---|
| Start planning | 100% |
| Find forecast | 100% |
| Review outlook | 80% |
| Find remarks | 80% |
| Open planning | 100% |
| Set field limit | 60% |
| Set facility limit | 100% |
| Change line status | 40% |
| Change facility status | 80% |
| Measure | Ease rating |
|---|---|
| Start planning | 4.8 / 5 |
| Find forecast | 4.6 / 5 |
| Review outlook | 4.8 / 5 |
| Find remarks | 4.2 / 5 |
| Open planning | 5 / 5 |
| Set field limit | 4.2 / 5 |
| Set facility limit | 4.6 / 5 |
| Change line status | 4.2 / 5 |
| Change facility status | 4.2 / 5 |
The line-status task recorded 40% completion alongside an ease rating of 4.2 out of 5. The observation notes explain that participants looked in a different planning area. This is why ratings alone would have concealed an important navigation problem.
Participants tried to change a default value rather than using the add-limit action.
Users looked for an operational status change in another planning tab.
Review tasks depended on finding comments alongside forecast data.
The most useful finding is the one that changes what you build next.
Real people. Real tasks. Evidence that gives the next design decision a reason.
Moderated usability report, pp. 2–5. Charts transcribe the key-results table. Labels are generalised. Five participants is a small sample, not a population estimate.
Task-level results identified specific discoverability problems that broad satisfaction ratings would have missed.
Tell us about your product, your questions or the skills you want to build.