What should you fix when someone opens your small tool but does not use it?
Find the first observable blockage before rebuilding the product. Distinguish unclear purpose, difficulty starting and an unhelpful result. Each stopping point suggests a different response.
Observe one real attempt
- Ask permission to watch.
- Offer a small relevant task.
- Record the first blockage without taking over. If a reading-list tool hides its add button, that alone says little about whether its recommendations are useful.
Repair the nearest verified problem
Ask what the person was trying to do and separate their wording from your interpretation. After fixing the entry point, retry the same task rather than changing the name, layout and features together. If the person has no relevant need, reconsider participant fit instead of insisting on repeated attempts.
What to do with your observation
| Observation | Next action |
|---|---|
| Purpose is unclear | Clarify the task and description before adding features. |
| Entry works but need is absent | Revisit the intended situation; a new button alone cannot resolve this. |
A worksheet you can use
Task: __. First blockage: __. Participant account: __. Interpretation: __. Single change: __. Retry: __.
Must I record video?
No. Brief notes may suffice. Recording needs separate permission and should avoid private screens.
Evidence and scope
This original observation workflow is not a large usability study. One participant cannot represent all users.
Graham discusses early manual work. These procedures are original. Paul Graham (2013) — Do Things that Don’t Scale
Sources and further reading
This collection in Chinese and English · Life possibilities library