I am starting in UX and I am little bit confused (already) about "Empathy Maps". I read this kind of article that explains the concept. How come can you know what a customer or user - for example - "does" if you're in an interview or using a survey? I mean, it's not a usability study or test where you'd normally be present. Using surveys, you cannot know what they "think, does and/or feels".

It's probably a bit silly but I'm really confused.

3 Answers 3


With an interview, the best you can do is simply ask them what they would think/feel/do in a particular situation. For example, you can ask them what they would do in a particular situation. Then you can ask them what they would be thinking in that situation and how it would make them feel. I think the hardest differentiation would be between think/say using this approach.

In a survey, you would have to take the same approach, asking opening ended situation based questions. You can attach images of a page to the survey and ask them what they think this page is for, what they would do on the page, how it makes them feel, etc.

Ideally, you would be able to give them a usability task which you can view. If you're doing this remotely then attach a survey afterwards asking them what their thoughts and feelings were during specific tasks.

  • This is a bit weird (I suppose because the fact that I am a newbie). It would make more sense if you are in a testing interview (so with your prototype for example) because the "thinking" or "does" thing is more an approach and it could be biased by your own interpretation and what the interviewe would think he/she has to say in order to "please" you or giving a "wrong" image, etc. Could this be the case where you have to think about another "approach" to empathize with your users? 🤔
    – ThisisUX
    May 29, 2022 at 9:27
  • 1
    It is always important to emphasize at the start that you are looking for their honest feedback. I usually say something like "please don’t worry that you’re going to hurt our feelings. We’re doing this to improve our product, so we need to hear your honest reactions." If one's feedback is overly positive and not saying anything constructive, then it might not be worth including that data. In a usability test, you can ask them to think out loud which gives you the best look into their internal dialogue. We are all subject to bias, so follow up questions can help clarify their actions. May 30, 2022 at 19:10

The way I would approach this in a usability test, “says” would be what a user says during the test without any prompts other than being asked to talk through what they are doing.

After the usability test, during the interview phase, we’ll often ask the user to further clarify “what were you thinking when you did/said this.” We also might ask how they felt during a particular task. Technically this is what they “say” but is does take more prompting, and you do have to be careful that your questions aren’t leading to a particular response.

The article you linked seems to imply feeling and thinking as giving more context to what the user says, this could include some interpretation based on tone and body language that aren’t necessarily apparent from the written quote, but there’s a slippery slope here to putting words in the user’s mouth, so I would personally avoid this in most cases. With general emotions, like frustration and boredom, this might be more acceptable, but I wouldn’t want to say that a user felt satisfaction, for instance, without the user specifically expressing that.

  • I think it has more sense (in U.testing) but I don't see it the same way when it comes out to surveys (exploratory surveys) since I could ask what they feel, hear, or does while they're answering the questions but I do not know what they could say about. "what do you hear?" (my dog breathing?, my washing machine?), or "what are you doing?" (writing, clicking?). I'm talking about when it comes to feel, hear, does, etc) is more about their everyday life? or when they do what I am researching? (for example, buying articles on internet). Maybe I am just missing the point.
    – ThisisUX
    May 29, 2022 at 20:45
  • I think, at the end of the day, it’s one way you can present your data, alongside personas, user journeys, mock ups, and others. If it doesn’t make sense for a particular project, there’s no point to pigeonholing your findings into this format!
    – TzeraFNX
    May 30, 2022 at 12:37

Research (in this case, empathy) and user testing are two different stages. Of course, you cannot be 100% sure that the user will tell you the truth. But if you build the interview correctly, you can get enough reliable data or a direction where you need to move. And then you can use other research tools.

The point is that during the interview you collect information about the user, and then its structure according to the Empathy Map

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.