Critical Incident Technique
A structured method for eliciting detailed firsthand accounts of specific events that materially helped or hindered an outcome.
Download Markdown (.md)Soroush’s Product Wiki · https://wiki.soroush.uk/topic/critical-incident-technique/
01What it means
The critical incident technique asks participants or observers to recall a concrete event, the circumstances and actions around it, and how it affected a defined positive or negative outcome. The incident is examined for requirements and contributing conditions rather than collected as a general opinion.[1][2]
- 01Define the activityOutcome, actors, and what makes an event critical
- 02Elicit one incidentA specific firsthand event rather than general opinion
- 03Reconstruct contextTrigger, environment, constraints, and intent
- 04Capture behaviorWhat people and systems actually did
- 05Trace consequenceWhy the event helped or hindered the outcome
- 06Classify patternsCompare incidents while preserving exceptions
An incident is selected because it materially helped or hindered an outcome; memory, selection, and interpretation limit generalization.
02Why it matters
Important failures or successes may be too rare to observe during a short study. Focusing recall on specific consequential events can uncover system, process, and service requirements that broad questions miss.
03When to use it
Use it to investigate uncommon, memorable, or consequential service events across a longer timeframe. Prefer direct observation or logs for routine behavior and verify accounts when decisions require causal or precise operational evidence.
04How it works
- Define the outcome and what qualifies as a relevant positive or negative incident.
- Recruit people with firsthand experience and ask for one specific event.
- Probe the situation, sequence, actions, people, tools, consequences, and recovery without supplying an explanation.
- Code incidents for recurring behaviors, conditions, and requirements while preserving exceptions.
- Compare with observational or operational evidence and translate patterns into decisions or further questions.
05A practical example
A team studying failed enterprise onboarding asks administrators to describe a specific setup that was abandoned. Several incidents reveal that security approval arrived after temporary credentials expired, pointing to a coordination requirement rather than a form issue.
06Where it breaks
Recall is selective and reconstructive, vivid extremes can overshadow typical experience, and participants may wrongly infer causality. The method does not estimate frequency or replace observation of routine work.
Common mistakes
- Accepting general complaints instead of probing one concrete event.
- Defining “critical” so narrowly that only dramatic failures qualify.
- Counting recalled incidents as prevalence in the wider population.
07Sources
- The Critical Incident Technique in UXNielsen Norman Group · Professional guidanceThe Critical Incident Technique in UX
- The critical incident techniquePsychological Bulletin via PubMed · Original researchThe critical incident technique