UX Metrics Cheatsheet
A quick reference for the metrics most commonly used to measure user experience. Understand what each metric tells you, how to collect it, and when to use it.
Last updated: July 2025
Behavioral metrics
What users actually do
Task success rate
What it measures: Whether users can complete tasks.
Calculation: (Successful completions / Total attempts) × 100
Collection method: Usability testing, analytics event tracking
When to use:
- Evaluating task flow effectiveness
- Comparing design alternatives
- Establishing baseline for critical tasks
Benchmark: 78% is often cited as average; aim for 90%+ on critical tasks.
Task completion time
What it measures: How long it takes to complete a task.
Calculation: Time from task start to completion
Collection method: Usability testing (observed), analytics (for instrumented flows)
When to use:
- Measuring efficiency improvements
- Comparing before/after designs
- Identifying bottlenecks
Considerations: Time includes errors and recovery. Successful-only time may be more meaningful.
Error rate
What it measures: Frequency of user mistakes.
Calculation: (Number of errors / Total opportunities for error) × 100
Collection method: Usability testing observation, form analytics, error logging
When to use:
- Identifying confusing UI elements
- Evaluating form designs
- Measuring impact of changes
Types:
- Critical errors (prevent completion)
- Non-critical errors (cause delays but recoverable)
Findability / Success rate
What it measures: Whether users can locate information or features.
Calculation: (Users who found target / Total users) × 100
Collection method: Tree testing, usability testing, first-click testing
When to use:
- Evaluating navigation and information architecture
- Testing search effectiveness
- Validating content organization
Single metrics rarely tell the complete story. Task success + time + satisfaction together reveal whether users can do something, how efficiently, and how they feel about it.
Attitudinal metrics
What users think and feel
System Usability Scale (SUS)
What it measures: Perceived overall usability.
Calculation: 10-question survey, scored 0-100 (not a percentage)
Collection method: Post-task or post-study questionnaire
When to use:
- Benchmark usability over time
- Compare products or versions
- Quick standardized assessment
Benchmarks:
- Above 68 = above average
- Above 80 = good
- Above 90 = excellent
Questions include: Ease of use, complexity, consistency, learnability
Net Promoter Score (NPS)
What it measures: Likelihood to recommend.
Calculation: % Promoters (9-10) - % Detractors (0-6)
Collection method: Single survey question
When to use:
- Tracking overall satisfaction over time
- Comparing to competitors
- Executive communication (widely understood)
Limitations: Doesn't explain why users feel that way. Follow with "why" questions.
Customer Satisfaction (CSAT)
What it measures: Satisfaction with specific interactions.
Calculation: Typically average of 1-5 or 1-7 scale ratings
Collection method: Post-interaction survey
When to use:
- Evaluating specific features or touchpoints
- Support interaction quality
- Transaction experiences
Task-level satisfaction
What it measures: How easy users felt a task was.
Calculation: Single Ease Question (SEQ) on 1-7 scale
Collection method: Asked immediately after task
When to use:
- Correlating with behavioral metrics
- Identifying friction points
- Comparing task difficulty across features
Benchmark: 5.5+ is typically considered good
Engagement metrics
How users engage with the product
Adoption / Feature usage
What it measures: Whether features are being used.
Calculation: (Users who used feature / Total active users) × 100
Collection method: Analytics
When to use:
- Evaluating feature discovery
- Identifying unused features
- Measuring launch success
Retention / Return rate
What it measures: Whether users come back.
Calculation: (Users active in period N who were also active in period N-1) / (Users active in period N-1) × 100
Collection method: Analytics
When to use:
- Measuring sustained value
- Evaluating onboarding effectiveness
- Tracking long-term engagement
Cohort analysis: Track specific user groups over time for more insight.
Frequency of use
What it measures: How often users return.
Calculation: Sessions per user per time period
Collection method: Analytics
When to use:
- Understanding usage patterns
- Identifying power users
- Evaluating engagement initiatives
Time on task / page / session
What it measures: Duration of engagement.
Calculation: Varies by metric
Collection method: Analytics
Caution: Longer isn't always better. Long time might indicate engagement OR confusion. Context matters.
Outcome metrics
Business results of UX
Conversion rate
What it measures: Users who complete a desired action.
Calculation: (Conversions / Total visitors) × 100
Collection method: Analytics
When to use:
- E-commerce and sign-up flows
- Landing page optimization
- Funnel analysis
Abandonment rate
What it measures: Users who start but don't finish.
Calculation: (Incomplete processes / Started processes) × 100
Collection method: Analytics funnel tracking
When to use:
- Form and checkout optimization
- Identifying friction points
- Prioritizing improvements
Support contact rate
What it measures: How often users need help.
Calculation: (Support contacts / Total users or transactions) × 100
Collection method: Support ticketing system
When to use:
- Measuring self-service effectiveness
- Identifying confusing areas
- Cost of poor UX
Comparative metrics
Before/after, A/B
Change in metrics
For any metric, track:
- Baseline: Metric before change
- Post-change: Metric after change
- % Change:
((New - Old) / Old) × 100
Statistical significance
Ensure differences are real, not random variation:
- Adequate sample size
- Confidence level (typically 95%)
- Effect size worth pursuing
Choosing the right metrics
| Research question | Primary metrics |
|---|---|
| Can users do this? | Task success rate |
| How efficient is this? | Task time, error rate |
| Is this easy to learn? | Time on first vs. subsequent use |
| How do users feel? | SUS, satisfaction, NPS |
| Is this discoverable? | Findability, first-click |
| Are users engaged? | Retention, frequency, feature usage |
| Is this working for business? | Conversion, abandonment |
Use multiple metrics together. If task success is high but satisfaction is low, something's wrong even though users "succeed." If satisfaction is high but conversion is low, there may be issues outside UX.
Quick reference card
| Metric | Type | Good for |
|---|---|---|
| Task success | Behavioral | Can they do it? |
| Task time | Behavioral | Efficiency |
| Error rate | Behavioral | Usability issues |
| SUS | Attitudinal | Overall usability perception |
| NPS | Attitudinal | Loyalty/satisfaction |
| SEQ | Attitudinal | Per-task satisfaction |
| Adoption | Engagement | Feature discovery |
| Retention | Engagement | Sustained value |
| Conversion | Outcome | Business results |
| Abandonment | Outcome | Friction identification |