Description
Rates each usability problem separately on impact, frequency and persistence, then combines them into a severity score, giving a defensible and reproducible prioritisation.
Nielsen defined usability problem severity as a combination of frequency (how many users meet it), impact (how hard it is to overcome) and persistence (whether users keep encountering it once learned), plus market impact. In practice teams assign a single gut-feel 0-4 rating, which correlates poorly between evaluators and favours dramatic over pervasive problems. The multidimensional approach rates each component on its own anchored scale, using observed data where possible (for example frequency as the proportion of participants affected), combines them with an explicit rule (multiplicative or a lookup table), and has at least two raters score independently before reconciling. This prevents high-frequency, low-impact problems from being dismissed and makes prioritisation transparent to stakeholders and comparable across studies.
Attributes
- Discipline
- Usability & Evaluation
- Loss type
- Latent
- Goal
- Prioritise
- Stage
- Evaluation
- Data
- Analytical / expert
- Effort
- Hours
Steps
- Rate each usability problem on three dimensions (1–5 each): impact (1=cosmetic, 5=task failure), frequency (1=rare, 5=universal), persistence (1=one screen, 5=systemic).
- Priority score = impact × frequency × persistence.
- This moves problems from anecdote ('this confused 3 users') to a quantified severity that can drive engineering prioritization decisions.
Sources
- Handbook of Usability Testing
- Usability Engineering (Jakob Nielsen)
- Sauro, Rating the severity of usability problems (MeasuringU)
Also known as
- Multidimensional severity rating
- Problem severity scoring
- Rubin criticality score
- Hall's severity-frequency tiers
Categories
- User Research
- Usability & Evaluation