How to run a usability test

Key plan for usability testing

Once we have something to test, we need to know how to test it. We can go out and test at any time, but a plan (even a quick plan) means we have a way to analyse the feedback we get more effectively,

There are a variety of user tests – but the one I use is adapted from the design thinking playbook (Lewrick, Link and Leifer, 2018).

User testing is broken up into four main stages: plan, do, document, analyse.

1. In the planning stage we decide what we want to learn and then we decide what we will test. This might mean we want feedback about the solution as a whole (is it desirable) or we may choose one aspect to test (will someone know how to navigate the webpage). Once we know what we are testing we then identify the user group we will test with. Target the appropriate user group to give the most relevant feedback.

   We should also plan the user testing scenario. i.e. what would be the best way to gather feedback. It is usually best to test more than one idea per session so that the user has the ability to compare and evaluate (which helps the user give feedback).

   By the end of the test we should know what to keep, what we should change and what we should discard with each concept/prototype/idea.
2. When we carry out the test we go in with a learning mindset. To help with this it is best not to test alone.

   Common roles are the moderator and the observer (or observers). The moderator sets the context and asks the questions while the observer watches in a focussed way everything the user does.

   It may be beneficial to record the user as it can be hard to note down everything that the user may (or may not) do. We could also take photos during or at the end of the test to capture different parts of the users experience with the potential solution.
3. The test is no use at all if documentation hasn’t been taken at the time. As mentioned video and photos are useful as ways to document. As well as the written observations and the written feedback of the user. This means we need to be able to capture and organise this data effectively.

   We use different feedback tools to capture the feedback and to prompt us to ask those deeper questions to uncover the motivations, needs of the user i.e. “Why did you do it this way?”, “Can you say more how this feels to you”.
4. Finally we analyse the feedback. Through this analysis we should be able to construct a stronger next iteration of our potential concept.

We then alter our solution accordingly, and repeat these 4 steps finetuning our concept, until we are satisfied that our solution is robust.

What could be missing from the high level plan above?

Post your thoughts in the comments below.