Description
Rates each usability problem 0-4 based on frequency, impact and persistence, by independent raters, so findings can be prioritised defensibly.
The treatise identifies the severity rating scale as a lost part of Nielsen's framework. Severity combines three factors: frequency (how many users encounter it), impact (how hard it is to overcome) and persistence (one-time or recurring), plus market impact. Ratings run from 0 (not a problem) to 4 (usability catastrophe, imperative to fix before release). Nielsen showed single-rater severity judgements are unreliable and recommended averaging ratings from at least three independent evaluators, collected after aggregation of the problem list. Without severity classification, organisations cannot prioritise findings, and reports become flat lists that are easy to ignore.
Attributes
- Discipline
- Usability & Evaluation
- Loss type
- Vestigial
- Goal
- Prioritise
- Stage
- Evaluation
- Data
- Analytical / expert
- Effort
- Hours
Steps
- Compile an aggregated, deduplicated problem list with descriptions and evidence.
- Send the list to at least three evaluators who rate independently.
- Rate each problem on frequency, impact and persistence, then assign 0-4 overall.
- Average ratings and check agreement; discuss large disagreements.
- Sort problems by severity and map to the release plan.
- Track severity distributions across releases.
Sources
- Lost Methods in UX — A Complete Treatise (Gopinath)
- Nielsen J (1993) Usability Engineering
- Nielsen J (1994) Severity ratings for usability problems (NN/g)
Also known as
- Usability problem severity rating
- 0-4 severity scale
Categories
- User Research
- Usability & Evaluation