I have a bit or a moral dilemma! While trying to recruit for some user testing, an opportunity arose to perform a diary study with a new user of our application. Seemed to be an ideal way to capture the experiences of a first time user.

Unfortunately, she's really struggling (to the point of anger), and it feels almost cruel to let her continue in this current state. Should I offer some guidance to help her out, or should I take the David Attenborough approach, and merely observe (and then offer her all the help she needs at the end of the study) ?

Would love to hear from others who've experienced similar moral dilemmas in the world of UX.

  • I don't understand how is this question about UX? Your question belongs at Workplace not UX. – Ivan Venediktov Nov 28 '17 at 11:32
  • 3
    The question is about a user-research method, and user research is vital to UX Design. So I think it should stay. – Ken Mohnkern Nov 28 '17 at 13:58

Yes, been in this situation where the user is just simply stuck and cannot progress. In this case you have to jump in and break the task and get the user to proceed anyway.

You have recorded the fact the user has encountered a catastrophic problem and recorded as such. Prolonging the agony yields no further useful data for you.

Edit: this is for an observational test; yours is slightly different. I still think if she can't continue at all there's nothing more she can tell you about the system; advising her how to get past the roadblock may get her commenting on other parts of the system she never would've pre-roadblock.

  • Maybe I should send an email of encouragement, and let her know that how much we're learning from her experience, and would help her out as much as we can after the study, but would appreciate her perseverance. For a bit of context - the application she's using is known for being quite intimidating for novice users - just didn't realise it was THIS intimidating ... – Michael M Nov 28 '17 at 22:21
  • well ok, you could try that but there is still a risk she will give up and at that point your learning from her is over. – colmcq Nov 29 '17 at 9:26

If the insights she offers you are valuable; continue. If it causes frustration, but she is unable to tell you why or the feedback is unusable; discontinue.

If her points are valid you can use this case to help prevent anyone else from experiencing the same problems. It'd be a positive experience for her if she can see that the issues she had were addressed and it wasn't her fault she was struggling. Discontinuing in that case would make her feel even worse probably ("It's my fault it went bad"). If the issues are relatively user specific (personal stuff and such), then you're not going to be able to leverage the learnings and it's a waste of time and an unneccessary emotional drain for her.

  • I agree with your point about maker feel like it's her fault, and exacerbating things by intervening. I mentioned in an answer above about possibly sending some words of encouragement. – Michael M Nov 28 '17 at 22:23

I think you should continue until the end before helping her out in between so that you can exactly understand why she is struggling.

Morally it's cruel to do this, but it will give you an opportunity to resolve the issue which might end up many other users struggle.


Is the cause of your candidate's struggle going to endanger them either physically or mentally in any way - either actively or by neglect?

Of course you have a duty of care to your candidates but you also need to discover the problems with either your product or your study design and you can only do that properly if the test runs it's course.

If, however, you are placing your candidate in jeopardy, then you need to seek expert advice regarding both stopping the study (removal of a tool in use may make the candidate's situation worse) AND a suitable recovery programme for the candidate (which should be covered by your company's public liability insurance).

If you do decide to stop the study then you also need to bear in mind the fact that you may have invalidated the result depending on how your study was constructed.

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.