Usability Testing

Watching Beats Asking

Usability testing means watching real people try to complete real tasks with your design. You are not asking whether they like it. You are observing whether they can use it.

It is the most reliable way to find problems before engineers spend weeks building the wrong thing. Research by the Nielsen Norman Group suggests that five users reveal most of the major usability problems in a design. Test with five, fix what you find, then test again.

---

Moderated vs Unmoderated

Many teams combine both: moderated sessions to understand problems, unmoderated tests to measure the fix at scale.

---

Recruiting Participants

Test with people who match your personas, not with your team.

• Screener questions filter the right people: "Have you sent money to a family member in the last month?"
• In Nigeria: recruit through community WhatsApp groups, university networks, market associations, or POS agents who meet your target users every day. Offer an incentive such as airtime, data or a small cash transfer (₦3,000 to ₦10,000 is common for a 30-minute session).
• Globally: platforms like User Interviews, Respondent and Prolific.
• Avoid: colleagues, friends, and people who work in tech (unless tech workers are your users).

Always get consent, explain how the recording will be used, and make it clear that you are testing the design, not the person.

---

Writing Good Tasks

A task is a realistic scenario with a clear goal. It should not tell the user how to do it.

• ❌ "Click the Send button and enter ₦5,000." This is leading. It gives away the UI.
• ✅ "Your younger brother needs ₦5,000 for transport today. Use the app to send it to him."

Good tasks:

• Use the participant's words, not your interface labels
• Have a clear end point, so you know whether they succeeded
• Come in a natural order, from simple to complex

---

Running the Session

1. Introduce yourself and the purpose. Say: "There are no wrong answers. We are testing the design, not you."
2. Warm-up questions about their habits and context.
3. Tasks, one at a time. Ask them to think aloud.
4. Stay quiet. When they ask "What should I do?", turn it back: "What would you do if I weren't here?"
5. Follow-up questions after each task: "What did you expect to happen?"
6. Wrap up and thank them, and send the incentive promptly.

---

Analysing Results

For each task, record:

• Task success: completed, completed with difficulty, or failed
• Time on task
• Errors and hesitations, with timestamps
• Quotes that capture frustrations or delight

Then look for patterns. One person struggling might be noise. Three out of five struggling at the same step is a finding. Rate each issue by severity (critical, major, minor) and fix the critical ones first.

---

Try It in the Lab

The script builder on the right helps you write a complete moderated test script. It flags leading tasks, checks your screener, and generates a printable script with intro, consent, warm-up, tasks, follow-ups and wrap-up.