10 tips to make you a better interviewer

Last week, James Lang and I were really excited to announce the release of Researching UX: User Research.

The book is a practical guide looking at the craft, technique and processes involved in running a research project. It is aimed at anyone interested in user research - those that are entirely new to UX through to those that have been doing it for years.

Our lovely publisher has kindly allowed us to share some of our book. We’ve decided to share an excerpt from the chapter ‘Interviewing’. In this chapter, we show you how to run insightful user research sessions.

One of the things that we spent time thinking about was the impact of wording on the success of interviews. Here are our top ten tips on how to ask a question in a research session:

1. Clearly state the subject and the query

Think of a question as having two parts:

  • The subject (what the question’s about: football, ice cream, holidays)
  • The query (what you want to know about it: what’s your favourite? what was your most recent experience?)

On their own, neither is much use:

Subject: “Let’s talk about ice cream”

Query: “What’s your favourite flavour?”

Yet when we combine them, it’s clear both what we’re talking about and what we want to know about it:

"Let's talk about ice cream. What's your favourite flavour?"

Asking the subject first, then the query, makes it slightly easier to process the question. But you can introduce them in either order:

"What's your favourite flavour of ice cream?"

2. Don’t use leading questions: establish relevance instead

A question like “What’s your favourite flavour of ice cream?” introduces a new issue: with this structure, we’ve assumed that the participant has a favourite ice cream flavour, and therefore the query is relevant to them. We’ve asked a leading question. This is bad news, because the participant may feel under pressure to pretend to like ice cream, and invent a favourite flavour. That’s worse than no answer at all.

To avoid this problem, we should use a qualifying question to establish the relevance of our query:

"Do you like ice cream?"

“What’s your favourite flavour?”

3. Keep your question as short as it can be

It’s easy to ramble on about the subject and let it turn into a kind of introductory statement, as in the example below. Apart from wasting time and patience, it’s possible that the participant will forget what you were asking about!

Subject:"The next thing I’m going to ask you about is ice cream, which is something that we’ll talk about now, and then later on in the interviews, but in the meantime I want to ask you about it initially."

Query: "What's your favourite type?"

It’s best to avoid this kind of rambling, but if you do need use an introductory statement, always restate your subject immediately before or after the query:

Introductory statement:"The next thing I’m going to ask you about is ice cream. We’ll talk about it later on in the session too. But I've got some initial questions that I need to ask you first."

Query & subject reminder together: "So. What's your favourite ice cream flavour?"

4. Closed questions and open questions have complementary strengths and weaknesses

Both closed and open questions have their role, and it’s best to use them together.

  • Closed questions (e.g. “Do you like ice cream?” or “Which do you prefer, cupcakes or donuts?”) are easy to phrase and understand. They require less cognitive effort from the participant to answer. They’re better for putting people at ease than more challenging open questions. And as we saw above, they’re good for communicating the subject and establishing relevance.
  • Open questions (e.g. “What do you like about ice cream?” or “What’s your favourite flavour?”) do most of the real work. They give participants the liberty to explore and talk about the subject in their own way, revealing their thought processes, needs and preferences in the process. Their disadvantage is that they’re harder to answer, can be more intimidating and can be a cue to ramble for some people. Open questions tend to begin with words like ‘what’, ‘who’, ‘when’ or ‘how’. It’s generally best to avoid ‘why’, because it can lead to post-rationalised answers.

5. Look out for ‘value words’

There’s a certain kind of word or phrase that you should be particularly attentive for when a participant is speaking. We call these ‘value words’: adjectives or descriptions which appear meaningful, but are actually open to wide interpretation. Some examples: “corporate”, “cluttered”, “gritty”, “feels like the real thing”, “my kind of place”.

When you hear value words, you should do one of two things:

  • Follow up with a probe there and then. You do this by repeating back the exact phrase or word, and then asking an open clarification question:

    “You said the website felt corporate. What do you mean by ‘corporate’?”

    "You said the shop 'feels like my kind of place'. What makes it feel that way to you?"
  • It might not be the right moment to follow up. Make a note of the value words on a pad or post-it note, and keep it in reserve to return to later.

Value words are your doorway to real insight in the interview. Don’t waste them!

6. Listen for contradictions and gaps

Imagine you’re a detective. The participant will be wanting to present a certain version of him or herself to you: perhaps someone who exercises every day and eats plenty of fruit and vegetables. Whether or not this is a conscious effort on their part, it’s unlikely to be the whole picture.

If you can be attentive to contradictions in the research session, you’re more likely to get under the skin of the topic, and find out what’s really going on. This might be a contradiction between what someone said and what they did:

"Earlier on you said that your preferred holiday destination was Mexico, but when you went onto the website, the first page you visited was South Africa. That's interesting isn't it? Tell me more about that.".

Or it might be an inconsistency between two different statements that they made:

"Earlier on you said that your favourite ice cream flavour was rum and raisin, but just now you said it was coconut. Do you have more than one favourite flavour, or is there another factor at play here?“

When you’ve spotted the contradiction, follow it up. It’s absolutely essential that you’re respectful in your phrasing, but participants are generally curious about their own internal processes, and intrigued by what you’ve shown them.

Sometimes it can be as interesting to note what’s not happened. Perhaps the participant had mentioned a subject they were interested in, but flicked straight past it in the material you were asking them to look at. Or perhaps you were expecting them to comment on a particular design, but they said nothing.

It’s ok to draw attention to this. Your expectations (your rolling hypotheses, in other words) need checking, so ask:

"A moment ago, I was wondering if you were going to click on to the information about Mexico, but you didn't. Looks like I was wrong! Can you help me to understand your thought process?”

7. Use question chains

Now that you’ve been introduced to the main question types, you can learn to use them in sequence, as ‘question chains’. Here’s an example:

Subject: "Let’s talk about going on holiday"

Relevance:_ "Have you been away this summer?"_

Query: "Where to?"

Participant responds: Sri Lanka

Open question: "What drew you to Sri Lanka?"

Participant responds: It was a great deal, and it’s a bit unusual

Probe: "When you say ‘unusual’…?"

Participant responds: I like to go to places that are a bit unusual. Somewhere a bit different. Not just the usual places that everybody goes to. And it's a resort, so you've got your home comforts too, there's a bit of predictability

Query contradiction: "Let me check then. So you're after somewhere that's a bit unusual, a bit different. And you also want your predictability, your home comforts?"

Participant responds: I hadn't really thought of it like that. I suppose what I want is a taste of adventure, without really going off the beaten track. Something to show off about at the water cooler when I get back to the office.

In that example, we’ve gone from using closed questions to establish relevance and initial facts, before using an open question to really kick off the conversation. We jumped on the value word ‘unusual’, and then explored the contradiction between adventure and home comforts. And in doing so, the participant has shared something that really matters to them: it’s important to them to be able to show off about their holiday afterwards. In other words, we’ve gone from some very neutral facts, to a statement about feelings and self-perception.

Once your chain runs out of steam, you can start another one. As you get more practiced, you’ll be able to run several question chains at once, spotting the contradictions between them as you go, and reviving them when the time is right.

8. Prompt, but don’t lead

When you want the participant to keep talking, but you don’t want to ask a new question, you can use prompts. It’s helpful to have a repertoire to use. These can include:

  • Short phrases such as “OK”, “Tell me more” or “Carry on”. It’s important to keep these neutral so that they don’t lead the participant: if you were to say “yes!” or “that’s right”, then you’d be tempting them to tell you more about what you wanted to hear.
  • Non-word noises like “um-hm” or “uh-huh”. These are better because they’re less of an interruption to the participant’s flow. On the other hand, they can feel awkward, so experiment to find one you’re comfortable with.
  • Body language. A nodding head can lead participants in the same way as “yes”, so go for something more subtle. One way is to tap your foot in time to the rhythm of their speech, and then keep it going if they stall. Or raise your eyebrows to indicate ‘go on’.
  • Silence is a great prompt (see below).

9. Choose the right moment

The timing of your question is as important as the phrasing. If you choose the right moment, you’ll barely have to give any explanation: the participant will understand what you’re interested in and give you a long, detailed answer. Choose the wrong moment and your question will fail, or get a radically different response.

Here’s an example. Imagine you wanted to get feedback about a new website design from your participant. Your plan is to ask them

"I'd like to know what changes you'd like the designers to make to this website. What are your top three suggestions?"

If you asked this question after the participant had used the site for just 60 seconds, then you’d be likely to get comments that focused on visual design: ‘the colours are too garish’, ‘I don’t like the photography’. If, on the other hand, you asked them at the end of the session, they might concentrate on difficulties with completing tasks, or missing functionality. Be flexible about the order you tackle subjects in, and don’t feel obliged to follow the sequence in the discussion guide.

10. Be quiet

Most people who are learning to interview talk too much. They don’t give the participant enough space to talk or think, and as a result they miss valuable insights. Get used to silence - become comfortable with it. Some participants (particularly more introverted personalities) appreciate a moment to gather their thoughts before or after answering a question. What’s more, silence is a great prompt.

Final thoughts

As you get more accustomed to interviewing, you’ll create your own methods and favourite questions. You’ll develop your own style. And because you’re comfortable with the technical side of interviewing, you’ll be able to pay deep attention at the same time as juggling your various other tasks.

Once you’re up and running, interviewing participants can be fascinating, and great fun. Enjoy it!

If you’ve found this useful, check out the rest of our book.

Emma is a User Experience consultant and digging through prototypes with users makes her happy. Recovering scientist, no cat, 1 gecko, can make fire without a match.