Usability testing is the cheapest way to find out why your page isn't converting. Five people trying to complete a real task reveal problems no report shows, because analytics tells you where someone gave up, but not why.
What usability testing is (and isn't)
It's watching someone who matches your customer profile carry out a specific task on your site, unassisted. It isn't asking for opinions, it isn't showing a layout and asking if they like it, and it isn't an internal approval meeting. Opinion predicts behavior poorly; observed behavior doesn't.
Formats and when to use each
| Format | Effort | When to use it |
|---|---|---|
| Moderated, in person or on a call | high | complex flows, understanding reasoning |
| Unmoderated remote | medium | validating simple tasks at scale |
| 5-second test | low | clarity of the above-the-fold promise |
| First-click test | low | finding out if the path is obvious |
| Heuristic evaluation | low | initial screening before involving users |
How to build the script
- Define a research question: what exactly do you want to find out?
- Write tasks with a scenario and a goal, never with click-by-click instructions.
- Recruit participants who match the real customer profile, not teammates.
- Ask for a think-aloud narration during the task.
- Stay silent: only step in when the person is completely stuck.
- Record where hesitation, errors, re-reading, and drop-off happened.
Example of a well-written task: "You want to know if this service fits a store with 30 products. Find that out on the site and, if it makes sense, request a contact." A bad example: "Click on Services, then on Request a diagnosis."
How to turn observation into priority
- Classify each problem by severity: blocks, delays, or annoys.
- Count how many participants ran into the same problem.
- Estimate the fix effort and prioritize high severity with low effort.
- Turn each finding into a testable hypothesis, with an expected metric.
These hypotheses feed directly into the CRO cycle and turn into experiments in the guide to A/B testing.
Combining with quantitative data
Before the test
Use heatmaps and session recordings to pick the page with the biggest drop-off and design tasks around it.
After the test
Measure the impact of fixes through the events set up in GA4. A fixed usability problem usually shows up first in the completion rate of your form.
Usability testing checklist
- Research question written down before the script.
- Tasks with a realistic scenario and no click-by-click instructions.
- Five participants matching the real profile per round.
- Sessions recorded with consent.
- Observation done in silence, without leading the answer.
- Findings classified by severity and frequency.
- Fixes turned into hypotheses with a success metric.
Conclusion
A half-dozen well-run sessions usually teach you more than months of internal opinion. Testing with real people is the shortest path between the page you think is good and the page that actually converts.
Frequently asked questions about usability testing
How many people do you need for usability testing?
Five participants per profile reveal most of the serious problems. Running several small sessions with fixes in between yields more than a single big round.
What's the difference between usability testing and A/B testing?
Usability testing is qualitative and explains why the user fails. A/B testing is quantitative and measures which version converts more. One generates the hypothesis, the other proves it.
Can I test with coworkers?
It's not recommended. Someone who knows the product overlooks obstacles that trip up a first-time visitor, which creates a false sense of clarity.
How do you write a good test task?
Describe a scenario and a goal, never the path. Instead of telling someone to click a button, ask them to figure out on their own whether the service meets their need.
Does usability testing require paid tools?
No. A video call with screen sharing and recording, with consent, is enough for the first rounds.
About the author
Yara Iencius
Partner and Head of Design & UX, KCLOCK
Conversion designer. Responsible for UX/UI, interfaces, landing pages, and e-commerce sites that turn strategy into a high-converting experience.
- #ux
- #research
- #testing
Want to apply this to your funnel with people who do it every day?
KCLOCK builds high-converting funnels, landing pages, websites, and e-commerce stores. We diagnose your current funnel and show you where the revenue you're leaving on the table is hiding.
