You have researched, structured, wireframed, prototyped and styled your design. Now comes the most honest step: watching real people try to use it. Usability testing means giving representative users realistic tasks and observing where they succeed, hesitate or fail. It closes the loop of the design thinking process and almost always sends you back to improve something.
Types of usability testing
| Type | How it works | Good for |
|---|---|---|
| Moderated, in person | You sit with the participant and guide the session | Rich detail, follow-up questions, testing on real phones |
| Moderated, remote | Video call with screen sharing | Reaching users in other cities |
| Unmoderated, remote | Participants follow written tasks alone using a testing tool; sessions are recorded | Quick results from more people; simple tasks |
| Guerrilla | Short, informal tests with people nearby, such as at a college canteen | Very early ideas, tight budgets – only if they match your users |
How many participants?
Jakob Nielsen’s widely cited guidance is that testing with about five users per round reveals most of the common problems in a design, and that several small rounds with fixes in between teach you more than one large test. If your product has very different user groups – for example, travellers and bus operators – test a few people from each group.
Planning the test
- Set goals. What do you need to learn? Example: “Can first-time users book a seat and find their boarding point?”
- Recruit. Find people who match your personas, not just your colleagues.
- Write tasks. Realistic scenarios with a clear end, not instructions.
- Prepare the prototype. Make sure the Figma prototype covers every path the tasks need.
- Run a pilot. Test the test with one person to catch confusing tasks or broken links.
- Decide what to measure. Task success, time, errors and comments.
Writing good tasks
| Weak task | Better task |
|---|---|
| Click the Search button and then select a bus. | You need to travel from Chennai to Madurai this Friday night. Find a bus that leaves after 9 pm. |
| Use the filter to choose Sleeper. | You want to sleep during the journey. Choose a suitable bus. |
| Go to My Bookings. | You want to check where to board the bus you booked. Find that information. |
Watch out: Do not use the exact words from the interface in your tasks. If the task says “Boarding point” and the button says “Boarding point”, people simply match the words, and you learn nothing about whether the label is clear.
Running the session
- Explain that you are testing the design, not the person, and that there are no wrong answers.
- Ask participants to think aloud – say what they are looking at, thinking and expecting.
- Stay neutral. If asked “Should I tap here?”, reply “What would you do if I weren’t here?”
- Do not help unless they are completely stuck; note where that happened.
- Take notes on behaviour (what they did) separately from opinions (what they said).
- End with a few open questions and thank them.
Rating and fixing problems
After the sessions, list every problem and rate its severity so the team fixes the right things first. A common scale runs from 0 (not a problem) to 4 (usability catastrophe – must fix before release). Consider how many people hit the problem, how badly it blocked them, and whether they could recover.
Before testing: a heuristic review
You can catch obvious issues before involving users by reviewing your design against Jakob Nielsen’s 10 usability heuristics. This is called a heuristic evaluation; it supports testing but does not replace it.
| # | Heuristic |
|---|---|
| 1 | Visibility of system status |
| 2 | Match between the system and the real world |
| 3 | User control and freedom |
| 4 | Consistency and standards |
| 5 | Error prevention |
| 6 | Recognition rather than recall |
| 7 | Flexibility and efficiency of use |
| 8 | Aesthetic and minimalist design |
| 9 | Help users recognise, diagnose and recover from errors |
| 10 | Help and documentation |
Tip: The blog post Nielsen’s 10 usability heuristics explained gives an example of each heuristic. Also run a quick accessibility pass: contrast, text resizing, and touch targets of at least 24 × 24 CSS pixels (WCAG 2.2 criterion 2.5.8), though larger targets are easier to tap.
Where to go next
You have now completed the UI/UX track. Take your practice project, run one round of testing with five people, fix the top problems and test again. Document the whole journey – research, personas, structure, wireframes, prototype, visual design and test results – as a case study. That story of decisions and evidence is exactly what a good portfolio shows. To start the cycle again, revisit the introduction or user research.