Glossary
Usability testing
What is usability testing?
Usability testing asks representative users to attempt realistic tasks with a product while researchers observe what happens. It can reveal where people become confused, slow down, or fail to complete a task.
How it works
Tests can be moderated or unmoderated, remote or in person. Formative testing looks for problems to improve; summative testing measures performance. Define the user goals, recruit suitable participants, observe attempts, and compare patterns.
How many users are needed?
A small iterative study can reveal useful problems. The familiar claim that five users find 85% of usability problems is a model-based expectation with assumptions about discovery. It is not a guarantee for every population, task, or system, or a sample-size rule for quantitative comparisons.
Example
Several users might struggle to apply filters while comparing plans. Watching their attempts could reveal a control they do not notice or labels they interpret differently from the designers. Investigate the difficulty and test a change. The observation alone does not establish its frequency across all users.
Why it matters
Seeing someone attempt a task can expose a mistaken design assumption quickly. Keep the findings tied to the people, tasks, and conditions tested, and combine them with other evidence where needed.