I'm creating a customer survey to collect feedback about our user experience and user interface. As part of the survey, I want to try to determine what works, what doesn't work, and what is missing. What is the best way of asking these questions in a survey?

Some suggestions that I've found include:

  • What features could you not live without?
  • What features could you live without?
  • What do you find the most frustrating about our application? (gets more at pain points but could be really useful to know)
  • Is there anything that you wish our application allowed you to do that it doesn't allow now?

Are these the questions you would use? Are there others you recommend?

  • How big is the interface? That is, let's assume that participants feel motivated to do the hard work of scanning their memories for stuff relating to this interface, when it comes to living with and without features, how many different features would you anticipate people talking about? 5? 15? 50?
    – dennislees
    Oct 13, 2015 at 19:01

3 Answers 3


There are a couple of suggestions if you want results you can trust.

Avoid confusing language: the first two questions are very similar. I'd suggest changing them to "What features do you consider vital to this product?" and "what features could you live without?"

Also, avoid 'personalising' language. Using words like "our product" will lead the respondent to worry that they will hurt someone's feelings if they offer a negative answer.


Easiest way I've found is to send it out to a sample group of the database first, see what feedback you get and how they answered the questions.

You might find people answering the survey in unexpected ways, or it may highlight some confusion in the way a question is asked. You can therefore adapt your survey and go on to send it out to the rest of the database.

Good luck!


I wouldn't try to answer these questions in a survey.

Looking at a product through the lens of features is a very product-centric viewpoint. Forcing a user to translate from their experience to a product-centric view loses something in the translation.

Users might be able to identify the features that they find most useful. Knowing that these features are the most useful doesn't tell you why they are the most useful, nor whether there are any improvements that you could make which would result in them being even more useful.

Users might be able to identify some features that they don't find useful. If they haven't discovered a feature, does that mean that the feature is not useful or simply well-hidden? If a feature isn't useful, is that because the feature isn't necessary or because its design needs to be improved so that it meets your user's needs and expectations?

Users might be able to identify things that they find frustrating. They're most likely to either remember a single very bad experience, or their most recent bad experience. Neither of these is necessarily the bad experience that you just fix first.

With no context, users often have a difficult time answering the question of what they would like to do with a product.

The best way to answer questions like these depends on what your goals are and what resources you have available to answer your questions. Most of these are likely best suited to qualitative feedback, such as an interview or ethnography, than quantitative feedback like a survey. If you have some ideas about workflows in your product that might need improvement, conducting a usability study of those workflows will identify those places. If you want to prioritize improvements to existing features and new features in future releases, starting with a workflow analysis to understand the most common workflows and what works (or doesn't) will help you understand where to focus your efforts. If you're not sure where to get started, a short 5-10 minute interview with users will help you get your bearings. If I'm truly stuck, I will ask users "what do you think of [the product]?" and listen for both positive and negative experiences, and perhaps ask additional probing questions.

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.