UX & usability research as task based testing?

Question

is task based testing a good way to gather user's opinions and research UX and usability of app? By task based testing I mean we have app for certain type of users, so we gather few people that fit in this target group and we give them set of tasks to complete inside the app - without explaining them this app, just like normal user would use it. Tasks are created based on how user would use this app. So e.g. if it's shopping app then one of the tasks would be to find product X, Y and Z from different categories, add them to cart, select delivery address and complete purchase with specified payment method. Then I would check if they were able to complete these tasks (maybe they couldn't find a product) and ask for their opinion on each step of the flow. Was it easy to find product X? Did you see there's search and you can search in specified categories? Or stuff like that.

Is that a good way to do it? Are there better ways to do it? Are there any specific things I should pay attention to? Should I also record screen to get full insight of what they did or that won't give me more info than what I already can gather by asking questions?

Sorry for noobish questions, but I'm not an UX specialist. I never did such research. I tried to google it, but mainly found surveys and interviews as a way to gather user feedbacks.

Answer 1

There are a few good "getting started with usability testing" hubs around the Web, I like the one on usability.gov. A good testing process is:

1. Define your research question. Decide what you want to know. Example: "Will retail checkout clerks over the age of 55 be able to figure out what to do next when a customer's credit card declines on their terminal?"
2. Define your research method. The usability hub can guide you on which test is right for your questions.

3. Test with 5-12 people who fit a good representation of your user (but if they're hard to recruit, testing a broader population is usually fine; do make sure that they are over the age of 18. Testing children has some different considerations.)
4. Capture the tasks that they easily succeeded, eventually succeeded, and failed.

• A matrix in an Excel sheet is usually good, you can put Task1, Task2, Task3 on one axis and User1, User2, User3 on the other with their results in the middle (red, yellow, green cells), etc.

5. Write up some insights on that can be shared with the team that needs to act on them.

Answer 2

I think what you are describing here is testing a specific user flow. By giving all your users the same task, you will be able to see if you see recurring issues or themes. This will help you understand better what needs to be fixed.

if you don't give them a specific task each of your users may do different things on your app and have different feedback. While this might be good, it might not get the quantity of feedback on similar issues to justify if these issues are simply experienced by 1 or 2 people or many.