I am designing a product that is going to have many functionalities.

I don't have the time to make a complete interactive prototype of all the sections, but I want to start making specific A/B tests for several task-flows.

The question I am asking is if you have any advice on how to carry out these tests without an interactive prototype itself.

I had thought of showing the screens in Figma to the users, in individual sessions, tell them the task, and ask them what they would do on the screen, being myself the one to show the result of their action (showing them the corresponding screen) according to each version of the design.

But this is the first time I do it, and I don't really know how to do this, and I'm afraid that their answers will be conditioned by the session itself (having to answer quickly, with me there, not being able to investigate each link on their own, as they would do with a final product, seeing each section of the platform, etc.).

Do you have any method for these cases?

In addition, we also have a part of the project that has already been developed in code, which I think could be improved. How could I show users the developed version (which would have to be tested in some way) and the alternative, which are only frames in Figma, and would have to be tested in another way, to get answers through the test?


1 Answer 1


It's absolutely feasible to do testing without interactive prototypes. There are a couple of good tests you could run using mockups, or even wireframes:

  • Five second test: You show the user the screen for five seconds, then hide it. You ask the user to tell you their quick impressions and what they think they can do. This will test if you've sufficiently optimized the screen for the user's next step in a workflow.
  • First-click test. "A participant who clicks down the right path on the first click will complete their task successfully 87% of the time." You can show the mockup, tell the participant a little about the scenario and the task, and ask what they would be likely to do. If you can give them control of the mouse remotely, you can watch where they click. If not, they can verbally tell you what they would do, or where they would click.

It's great that you have live code to test, you can watch the user interact with that and watch for their challenges and struggles. You can also capture attitudinal scores - "On a scale of 1 to 10 with 10 being easiest, how easy was that? How useful is this feature?"

You do have to be careful comparing apples to oranges when you test a mockup vs a live system - for example, you wouldn't have the user do a timed task test clicking on both, then reporting that the mockup was X% faster.

In that case, you could get some benchmarking metrics by testing the live version with one group (capturing that it takes 20 seconds to complete the task, as an example) and then use that to understand the obstacles and what could be improved in the new design. You could then do first-click and five-second testing on the redesign with a second group and see if there are any improvements users would need. You can then build out the new user-informed design and do some true comparative testing once you have that set up in code.

  • Thanks for the detailed answer @Izquierdo Wants a Winter Hat
    – Lucio
    Commented Dec 9, 2022 at 14:04

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.