Follow a shopper from discovery to return information
Ask for a suitable product, a justified variant choice, a checkout review and the applicable return route. Record where the participant needs help rather than pointing at controls. The four-task sequence reveals whether product information and interaction support the purchase, while keeping any payment exercise inside an approved sandbox.
Worked example — illustrative data.
| Task prompt | Record | Proposed acceptance |
|---|---|---|
| Find a lamp for a narrow desk | First route, misunderstood labels, assistance | Reach a plausible product without coaching |
| Choose finish and confirm fit | Selected variant and specification used | Explain why the lamp fits |
| Review checkout without paying | Unexpected costs and unavailable states | Understand total and next step |
| Find how to return it | Policy route and unanswered question | Locate applicable instructions |
Use a fictional scenario or an approved test catalog. Give the goal without telling the participant which menu or button to use. Keep their observations separate from the facilitator’s interpretation. An incorrect choice can reveal missing product information even when every control works. For checkout, use a sandbox and stop before any real transaction. Save the device, browser, route and task wording so the next walkthrough can repeat the conditions.
Familiar staff can bypass unclear navigation
A participant who already knows the catalog may skip unclear navigation by memory. Include an unfamiliar shopper for discovery tasks, and label any employee walkthrough as an internal review.
Fix a demonstrated blocking defect first. Treat a single preference about colors or layout as a design question to investigate rather than a measured conversion opportunity.
Continue with Guides for stores, software teams and local businesses, Conversion audit for checkout and signup journeys and Website usability checklist for a buying journey.
For keyboard operation, see the W3C WCAG quick reference.