Task design determines which usability problems can appear
Aliases: task scenario · task coverage · scenario validity
What it is
Task-scenario validity depends on whether goals, starting states, data, constraints, and consequences represent target use. A test reveals problems near triggered paths. Omitting recovery, permission, boundary, or cross-channel states provides no evidence that they are usable.
Why it happens
Wording can reveal goals and navigation labels; prefilled data bypass input problems; ideal accounts avoid permissions. Task order adds learning and fatigue. Tasks therefore sample the product state space rather than neutrally expose it.
Studying it
Build a task library from fieldwork, telemetry, and support cases, labeled by frequency, consequence, segment, and dependency. Pilot for leaked labels and path cues. Balance order and starting state, and report uncovered paths.
Where it stops holding
Extreme tasks help explore risk but do not represent everyday frequency. Fully real tasks may be private, irreversible, or longitudinal and require safe substitutes with disclosed loss. Broad tasks cannot compensate for the wrong population.
Applying it
- Write tasks from user goals and triggers, not interface labels.
- Cover main paths, recovery, permissions, and critical boundaries.
- Randomize or balance order and note prior-task learning.
- Map findings to tested states and queue uncovered areas.