The best kind of user research data is contextual: you visit your users in their context and watch the way they work. But sometimes, contextual research isn’t possible. For example, I recently carried out research with a group of legal professionals and despite my best efforts, a large number of them would not allow me anywhere near their desk or working environment because of client confidentiality issues.

When you’re in the early phases of a project and this situation arises, how can you interview users and still gain some understanding of user needs?

Consider for a moment the kind of data you tend to collect in early phase user research. The most useful data tends to be rich descriptions — stories — of the way your users work. One reason why context is useful is because it helps people tell and remember their stories. Out of context, this becomes harder:

• The participant may not be able to think of any stories.
• The participant may start on a story but dry up quickly.
• Stories may be mundane and trivial.
• The interview may fizzle out and become just another opinion discussion.

Is there a way you can help your participant recollect stories?

The Episodic Interview is an approach to interviewing where participants are encouraged to recall concrete events, situations and episodes around an experience. I first came across this approach in a book chapter by Uwe Flick.

Flick describes 9 phases in an Episodic Interview and I’ve structured this article around her framework. Let’s look at each of these in turn to see if it addresses the issues I’ve listed above.

As I review each of these phases, I’ve applied the method to an imaginary project. Let’s assume we’re working with an organisation that’s creating a kind of “Uber for car pooling” service called “Rideshare”. The service is about filling empty seats in vehicles.

Phase 0: Preparation

In this phase, you prepare an interview guide just like you would with any other interviewing approach. So you would do some reading around the topic of car pooling, you’d look at any other user research that has been carried out, you would speak with stakeholders to understand their vision for this thing that’s being developed, and you would list the key participant characteristics and recruit people for the interviews. During the recruitment, it’s worth asking your participant for a relevant story, just to show that they do have something to tell.

Phase 1: Introducing the interview principle

In this phase, you explain the purpose of the interview to your participant: you are looking for them to recount stories. An opener like this should work for our project: “During our interview, I’ll ask you to tell me stories about your experiences with carpooling or lift sharing. I’m interested in any stories, situations and events that stand out to you.”

During the interview, you may need to return to this point, especially if you have a participant who provides short answers to your questions. You can re-orient your participant with prompts like, “Remember, I’m interested in stories about that experience”, “Tell me more about that in the form of a story”, or “Can you recount a story about that?”

Phase 2: The interviewee’s concept of the issue and his/her biography in relation to the issue

We now move on to get a high-level overview of the topic, starting with the participant’s definition of what the topic is about. A good question here would be: “What does lift sharing mean for you? What do you associate with the word ‘carpooling’?”

You’ll then move on to ask participants about how they first came across the thing you’re investigating. Remember that your main goal with this line of questioning is to encourage participants to recall specific episodes: stories, anecdotes and experiences that provide insight into their goals. So appropriate questions here would be:

• “When you think back, what was your first experience with car pooling? Could you tell me about that situation?”
• ”What was your most relevant experience with car pooling? Could you tell me about that situation?”

Note that word “relevant” in the second question. Relevance is always determined by the participant, not by the interviewer. By asking the participant to determine the most relevant experience, you are much more likely to address the underlying user need than if you ask them to list various experiences and then you decide to focus on one of those because you find it most interesting.

Phase 3: The meaning of the issue for the interviewee’s everyday life

In this phase of the interview, your goal is to encourage participants to consider the meaning and relevance of your research topic and to help them describe how it fits into their day-to-day lives.

A useful question here would be, “Could you please recount your day yesterday and where and when car pooling played a part in it?”

As participants recount situations, you’ll then follow up with appropriate questions. For example, assume that a participant talks about car pooling in the context of commuting to work. A follow up question would be, “If you look at the whole experience of getting to and from work, what role does car pooling play in it? Could you tell me about a relevant experience?”. If instead, the participant talked about car pooling as a way to travel to a sporting event with fellow fans, you would follow up with, “If you look at the whole experience of attending a football game, what role does car pooling play in it? Could you tell me about a relevant experience?”

Phase 4: Focusing the central parts of the issue under study

In this phase, you move from the general topic (car pooling) to the specific (using technology to help people share rides).

This is a difficult phase for you as the interviewer because it’s important that you avoid trying to validate a specific implementation idea. It may well be the case that everyone on your team has a very definite idea about this thing ending up as mobile app, but your task as a user researcher is to allow participants to tell their stories, not gain approval for your design concept.

So questions like these:

• ”If you look back, what was your first experience when organising a ride share? Could you please recount that situation for me?”
• “What do you do when you need a ride share? Tell me about a typical situation.”

These questions should elicit the technologies that people use (like text messaging, email, web sites), but if not, you could be more explicit: “Tell me how you use technology to organise a ride share.”

Flick makes the point that in this phase your role as interviewer is to “open the doors to the interviewee’s personal experiences”. This means you need to probe deeper and deeper to generate a fully fleshed out narrative.

Phase 5: More general topics referring to the issue under study

The earlier phases have been getting increasingly specific. In this phase, you take a step back and ask the participant to think more generally. For example, questions like:

• ”In your opinion, what would make the process of car pooling easier?”
• ”What developments do you hope for in the area of car pooling in the near future? Tell me a story that would make this change clear for me”.

The aim of these questions is not to ask participants to design your system but to uncover discrepancies and contradictions between their opinions and the personal experiences they described earlier. A gap between what people say and what people do is often a design opportunity.

Phase 6: Evaluation and small talk

In this phase, you wrap up the interview and ask the participant if there was anything you should have asked, but didn’t. This allows your participant to add other stories that may not have been top-of-mind in the main part of the interview.

Phase 7: Documentation

Immediately after the interview, you should summarise the main findings. I used to do that on a prepared sheet but more recently, I’ve found myself dictating the stand-out observations into a voice recorder. This is because I’m often travelling between participants, and this allows me to summarise the findings while grabbing a coffee or waiting for a train. I can usually summarise an interview in 5 minutes.

Phase 8: Analysis

Analysing the results from qualitative interviews like these is very different from analysing the quantitative data you get in summative usability tests. With quantitative data, means and variance are useful ways of summarising the data but there's no benefit in creating an 'average' story. Here's a general approach I use with qualitative data like this:

• Get familiar with your data. Get the interviews transcribed and read through the transcripts a few times.
• Identify significant stories: “What happened?”
• Use affinity sorting to group similar stories from different participants.
• Interpret the data: “What’s behind this cluster of stories? Why are they significant?”
• Develop a description. Summarise your data so the design team can take action on it.

The Episodic Interview is a specialised tool that you may not use often but that is extremely useful in particular situations. It provides a useful framework for understanding user needs when you're unable to observe users in context. The focus on stories and real examples generates rich data to help you empathise with your users' day-to-day lives.

Reference

Flick, U. (2000). Episodic interviewing. In 'Qualitative researching with text, image and sound: A Practical Handbook', pp75-92. Atkinson, P, Bauer, N. W., Gaskell, G. (Eds).

Acknowledgement

Thanks to Dr Philip Hodgson and Gret Higgins for their comments on this article.