I'm going to be running a task-based test with some members of staff for a new version of an internal tool that a few departments use daily.

I'd like to get the test candidates to tell me what their expectations are of the ideal workflow when using the application. I'm concerned that by simply asking them, they'll be biased towards how their existing software works and try to explain how they'd tweak that instead of a more holistic approach.

My colleague has suggested that I produce steps of a workflow on cards/post-its and get them to go through each one and arrange them in the order that they'd like to do them.

This sounds like a good idea, but I don't know what this is called in terms of usability tests, does this have a name?

Does anyone have any other methods that they've used to understand a user's ideal workflow?

  • What you describe could be called "Paper Prototyping" which was popular a few years ago. Really, I've found the best way to tackle this is to get the users to talk you through what they need, perhaps with some props like sketches of UI, a whiteboard, etc. – Steve Jones Aug 6 '15 at 16:16

You could probably adapt a card-sorting tool to find out what sequence most people expected to do the tasks. Or you could build a couple of prototypes with alternate task flows and test those. A quicker way might be to look for similar workflows on competitor sites and test those with a tool like http://www.usertesting.com/. Or, as Steve suggested above, sketch up some wireframes of your potential workflows and test those. For one-on-one tests like that, Silverback is a great tool.

| improve this answer | |
  • I wish Silverback was working :( 2.0 doesn't work with current Mac webcams and 3.0 was pulled due to bugs. – Phillip Quintero Aug 6 '15 at 19:33
  • You can try using a video conferencing tool like gotomeeting or webex to capture your session. It doesn't highlight screen clicks for you or capture user's facial expressions during the session. But on screen interactions and tone of voice is usually enough. – nightning Aug 6 '15 at 20:40
  • Seems ironic that a user testing tool doesn't work properly despite testing with over 100 beta testers! I'll be doing a task using a mock-up, which I'll capture using Camtasia, but first I'll do a task-flow analysis like @RobC has suggested to find out the user's ideal workflow. Something a simple as writing down steps on post-its and then arranging them into a flow chart should give me the information I need. Comparing the workflow to how they respond to the mock-up should hopefully show where changes should be made. – Ben Harvey Aug 7 '15 at 9:15

This seems like a candidate for a/b testing where you create different workflows and then get subjects to rate their experience. Whichever workflow scores higher will most likely be the one to proceed with. You can also reinforce your findings by a questionnaire of some sort afterwords.

| improve this answer | |
  • Adding a questionnaire afterwards is definitely something I should consider. – Ben Harvey Aug 7 '15 at 9:16
  • it worked wonders in my experience. it allowed us to pinpoint the major flaws in design and also users gave other ideas that we totally overlooked in development. – Stanley VM Aug 12 '15 at 21:30

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

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