Advanced UX interview techniques

One of the oldest and most frequently used methods to understand users, their needs, attitudes and behaviours is the interview. After you developed a questionnaire, you can use it verbatim and you will get consistent and structured answers. You can also use a semi-structured questionnaire, where you have the basic questions, but have the freedom to explore each previous answer with new questions that you create at the spot. You may have some negative consequences such as more time to conduct the interview and more work to analyse the responses.

One of the techniques for conducting semi-structured interviewing is ‘laddering’. It has this name because you take one answer from your interviewee and asks another question about what they just said, but with each new question going up a level, obtaining more details. You may eventually reach the core of what the interviewee has to say – the motivations and values behind their choices.

On the first levels of basic interview questions and answers, there are usually the general attributes (A) related to the product, service or anything you’re investigating. On the second level are the consequences (C), and on the third level, you may find the values (V) that users hold.

For example, imagine you’re doing UX research for a large company producing potato crisps and chips. Your client is considering adding another flavour to the range of crisps they already have. So you set up a tasting session providing a couple of flavours, maybe the regular ‘ready salted’ crisps, and the new strong flavour under investigation. You then invite people to taste them and give you feedback1. So your interviewee says:

I like the flavoured crisps

This is an interesting answer, and give you some information about the preferences. After a good number of participants, you can probably say that certain percentage of them prefer the new flavour, from a quantitative point of view. But you can explore a bit more. For example, you can ask why. Then the participant says:

It has a strong taste

Good, now you received one attribute (A) about the product, but this is probably obvious by the ingredients. You can probably say whether or not your participants are ‘getting’ the new product. But you can explore a bit more, going up the ladder, with other ‘why’. For example: why are the flavoured crisps better?

Because then I eat less

Now you’ve obtained the first consequence (C), as a development from the previous answers. Flavoured crisps may be preferred because then consumers will eat less. But if you stop here and report this as the conclusion from your research, the client may decide that the new flavour is not a good product. It may cause in fact a reduction in sales, a consequence that they probably don’t want to consider. Instead, you can explore a bit more, asking: Why eating less is preferable?

Because if I eat less I don’t get fat

Your interviewee gave you another consequence (C), which is fundamental to them. You notice that this participant is conscious of the high calorie of potato chips. They want to enjoy eating them, but if possible, avoiding negative consequences. You can carry on asking, ideally something carefully phrased. For example: why not getting fat is important to you?

Because I value being in a good shape

Here you reached the level of values (V), exposing what governs people’s decisions, based on the consequences of product attributes. If you feel there is room for further investigation, you can still explore a bit more, for example asking: Why you want to be in a good shape?

Because it affects my self-esteem

Here you probably reached the top of the ladder, when you reach the core of value, beliefs, attitudes, anything behind the psychological motivations of participants.

This information was only obtained by using the semi-structured interview and using the laddering technique. When you reach these deeper values, you have information to help create products or marketing strategies more adequate to specific audiences, based on what people really want. One ad campaign saying that these specific potato crisps are tasty perhaps won’t convince many people. But a campaign selling the lifestyle of a healthy life, with people in good shape and with high self-esteem who still eat this snack, probably will have a bit more success.

It’s important to highlight that this process should be part of a natural conversation, where you pose the questions as you go along, having in mind the main aims and objectives of the research, plus an explorer feeling and a curious mind.

But to execute this methodology it’s necessary to know the product or service in question, know the technique, rehearse and pilot the questions and possible answers, and be able to adapt quickly to what participants say. If the interviewer doesn’t know how to conduct this technique, may obtain inconsistent results, or worse, end up irritating or annoying the interviewee.


1 This example was originally suggested by:

Reynolds, Thomas J. Gutman, Jonathan. 1988. Laddering theory, method, analysis and Interpretation. Journal of Advertising Research 28: 11–31.