A while back I clocked up my 1000th interview. This got me thinking how much my approach has evolved over the years.
Interviews with customers / end users of products and services are often the foundation of my research.
In the earliest projects I’d work from a page or two of questions all lined up in advance, in the shape of a ‘script’, or discussion guide. These were questions I’d literally recite to each participant. Sometimes these had been contributed to, signed off by, or even provided by the client.
I’d been told I should ask the same questions to all participants to maintain consistency, but found it awkward to work to the script, and at times like I was only hearing half of the story from the subject.
Over time, I found the questions I asked in response to the answers revealed more than the questions on my script, so I developed a more conversational approach.
Sounds like a convenient way to take the effort and rigour out of the process, but it doesn’t make it any easier.
Whilst interviewing, you’re running a mental cache of what’s been said, where you need to take the conversation, how much time is left etc. …and all the while you’re trying to make the participant feel like the conversation is following natural twists and turns, rather than being steered by you, the interviewer.
There are plenty of techniques to learn in the craft of interviewing; building rapport, non-verbals, open ended questions, asking ‘the 5 whys’, repeating their words etc. In my book – USERPALOOZA – A Field Researcher’s Guide I cover the three types of question I use in every interview; starting points, prompts and qualifiers.
These techniques, combined with your curiosity will get you so far. …But they are not enough.
When clients ask (and they still do) “So, what are the questions you’ll be asking them” …
I explain:
When it comes to asking the right questions, there is no substitute for actually wanting to know the answer.
Instead of a script, I agree on a set of objectives with the team. This describes the ground we’d like to cover during the conversations and reads like a list of topics around which we’d like to learn.
Some of these might be framed as questions, but it’s far from being a ‘script’.
As an interviewer, you need to truly understand the context and objectives of your client / project sponsor:
It all starts with a set of questions to which I need the answer in my own head, before I begin planning the interviews…
- Where is the business and product at in the development process?
- Why is this the right time to conduct the study?
- Which aspects stakeholders agree / disagree on?
- What assumptions exist about the market, end user or value of the product to end users?
- How will the client measure market success for the product / service?
- How will the research be used, by whom?
- What design decisions do the team need to make based on the insights you uncover?
- Why are we including these types of participant in the study?
- Which areas does the team have enough insight about already?
This goes beyond the due diligence of taking the brief, scoping the study etc.
It’s a deep understanding of the business, product and design context and should be embedded in your curiosity.
The flow of the conversation and lines of questioning should all come naturally if you’ve built this level of empathy for your client’s position.
In the end it’s about user centred design – The user of the research is your client, so you need to understand your end users’ needs to be able to design the product (interview structure) to give them the best outcomes. In this case, rich and useful insights.
Nice! I often explain the differences between “questions you want answers to” and “questions you ask” – however, I still do produce an artifact – I call it the Platonic ideal – of questions we will ask. I think it’s such a great previz of the interview for me and for the clients; I don’t have a hard time setting expectations that most interviews will in fact look nothing like this, though.
PS – thanks for the book mention!
Hi Nick,
I enjoyed reading your blog on interviewing and want to add a little more (maybe) to what you have written.
A long time ago I discovered, as you have, that it has to be a conversation, that developing a script is always a waste of time as the interview never goes according to the plan, and nor should it. Why? Well to put it simply what you uncover along the way is always a surprise, a discovery, a revelation, at least it should be, as that is the purpose of having the interview in the first place. In reality when you have a script, the element of surprise and discovery is somewhat pre-determined by the thread you plan to follow and the questions you plan to ask. An interview plan makes it easier to miss the deep insight.
I find that the interview process works best when I start small scale. I will have a consistent starting point, a set first question such as “tell me about the last time you ———-” and a solid sense of what the purpose of the interview is. I test the approach on a few people first. I assess the outcomes of the conversations and adjust the approach as necessary. I refine and adapt small scale until I believe I am getting to the core of what is needed and then I move to a more in depth conversational exploration.
Experimentation, prototyping and testing are core parts of the design process, appropriately applied in this case to the discovery activity.
Skills I have learned over time include probing the responses I get; as you say, asking why five times. Where I am on my own, (not ideal) I draw pictures and make notes as I go. I reflect back what I have heard in the conversation, I especially clarify and confirm, the best way to build trust. I try to make greater use of my ears and eyes before my mouth. I always challenge assumptions both of my own and of the person being interviewed. I sometimes find it pays to take a pause, a deep breath if you like, to allow thoughts to settle and for the interviewee to think more about the subject of the interview than the question that has just been asked. Silence is golden!
At the end, I try to explore with the interviewee the insights I think I got; it is amazing how frequently I get a richer insight as an outcome.
Thanks again, I hope this helps.
Cheers,
John.
Wow, thanks John…
Thanks for sharing your experience.
Yes, I agree, Ears and Eyes before mouth.
If anything the ratio of 4:1 input devices (eyes/ears) to output (mouth) is stacked a little too much towards the interviewer talking rather than listening / watching.
If the interviewer is talking for 20% of the time it’s not going well!
Thanks Nick. Again you put things so well.
I’ve conducted hundreds of hours of interviews without ever writing a script and I’ve never had a complaint about the depth of insights I’ve uncovered nor did a session go badly.
However, I do have a loose framework that I create in Omnioutliner. These are theme buckets that I can write user’s comments when they bring up the associated topic. I create new ones and delete redundant ones as I go.
Since going freelance companies, especially the bureaucratic ones, always insist on a script. They even pour over the exact language used. Nobody reads the script, they just want to see pages and pages of questions and lots of detail. It gives them a sense of security that things are being done properly.
I’ll leave you with a quote from Albert Einstein… “If we knew what it was we were doing, it would not be called research, would it?”