ISTQB CT-UT study materials
ISTQB CT-UT study resources

ISTQB CT-UT: Usability Testing Methods and Approaches

ISTQB CT-UT · Work through your exam, one topic at a time.

All guide sections

Usability Testing Methods and Approaches

This domain tests whether you can choose a usability method that actually answers a given research question, and whether you know the limits of each method. Expect items that contrast expert inspection with user-based testing, formative with summative goals, and moderated with unmoderated sessions.

Two families of methods

  • Inspection (analytical) methods rely on experts, not users: heuristic evaluation against Nielsen's ten heuristics, cognitive walkthrough (four questions about whether a user will try the right action, notice it is available, associate it with the intended effect, and see progress), consistency and standards inspection.
  • User-based (empirical) methods observe representative users: moderated lab sessions, remote moderated or unmoderated sessions, think-aloud (concurrent or retrospective), co-discovery, field studies, interviews and questionnaires.

Formative versus summative

Formative testing happens during design to find and fix problems; it is qualitative and needs few participants per user group, often around five until findings saturate. Summative testing evaluates a finished or near-finished product against measurable criteria; it is quantitative and needs larger samples. ISO 9241-11 measures are effectiveness (task completion), efficiency (time, effort) and satisfaction.

Worked example

A team must prove that a redesigned checkout meets "at least 90% of users complete payment in under three minutes". That is a summative question, so heuristic evaluation is insufficient. Run a moderated, task-based test with participants matching the buyer profile, record completion, time on task, error count and a satisfaction questionnaire such as SUS, and report in ISO/IEC 25062 CIF format. Think-aloud is fine in the qualitative pilot but should be dropped for timed measurement because verbalising slows users.

Traps

Calling an expert review "usability testing"; using five participants to claim a quantitative benchmark; letting think-aloud inflate task times; writing tasks that name the menu path instead of the goal.