It’s my turn to cook tonight. I groan and reach for my quick recipe guide (as always). Time taken to prepare the meal? 15 minutes max.
My partner takes a totally different approach. He scours the internet for hours. Cycles to the Turkish grocery store to buy dates and goes to the local market for a freshly caught trout. Then he spends all day Saturday singing in the kitchen, as he cuts, stirs and tastes.
No points for guessing who prepares the most delicious meals in our household!
It’s the same with research. To get a good result you have to put your body and soul into it.
What makes UX research so complex?
As humans, we don’t have a clue why we do the things we do. Because it’s our subconscious that more or less determines our behaviour. Think back to the last time you visited a webshop and left without making a purchase. Why was that? Can you remember?
You might say: ‘They didn’t have what I was looking for’. Or, ‘It was too expensive’.
But who are you trying to kid? The truth is, there was something completely different going on. The website looked unreliable. The pages took too long to load. And it was difficult to find what you wanted. By the time you thought you were ready to make an order, your subconscious had already decided for you: get out of here.
But we consider ourselves to be rational beings. So, after the fact, we think up a good story to back up our decisions. And for ourselves. ‘I really need that new iPhone for my work. And then I can give my children the old one.’
So, OK. How do we obtain reliable information?
By taking the interviewing process seriously. If you want to gain reliable information from an interview, you have to know which questions to ask and which not to ask. And which answers you have to take seriously, and which not: that it does make sense to listen when someone says they don’t understand a particular word, but not liking the colours on a site is not relevant. Because the latter doesn’t say anything about how they navigate on your website. Take a look at Booking.com, for example: difficult to imagine a less attractive site, but it’s so effective.
When you’re interviewing, act as a detective:
- Listen and observe – Observe how participants treat your design. Don’t ask any questions, just watch what they do.
- Speak their language – If you start throwing jargon around, participants won’t feel comfortable. They might even feel stupid. If they feel intimidated, they won’t dare to admit that they really don’t understand anything you’re telling them. They’ll think: Well it must be down to me.
- Look for evidence – Go in search of evidence to back up what your participants are telling you. Ask for examples. If they claim to be extremely price-conscious when they go shopping, ask them how much money they spend on groceries every week. Get them to show you a receipt. And dive into the data: is the analysis you’ve made of the online behaviour of visitors in line with what the participants are telling you?
- Formulate hypotheses together – Once the interviews are completed, take stock of all the information in collaboration with your team. What did you notice? What did you hear? Think up hypotheses together based on the behaviour. We observe participant A hesitating about buying a dress; she looks at the photo for quite a long time. She says she wants to look somewhere else. We think the photo of the dress isn’t clear enough.
And does that qualify you to be a researcher?
You can ask everyone to do some programming but that doesn’t make them all programmers. They need to get at least a little excited about code. And feel the urge to keep up with the latest developments. The same applies to research.
Well, you might be thinking, what harm can it do if everyone in the team conducts an interview now and then? That way they can identify with the clients and understand them better
Maybe, but then you’ll end up making design decisions based on nonsense. Let’s take another look at the webshop. The test subject thought it was an attractive website, but the products were too expensive. Are you going to adapt the prices accordingly? But what if the real problem is the poor search functionality?
An example: extra functionality
I’m conducting user research for a new professional printer. You know, the massive ones you see in copy shops. In my preliminary research, I ask the participants what they think of it. They’re positive, but also have lots of ideas about making things even easier. For example, when the paper is about to run out, they want to see a message in the control panel. And they want to be able to schedule print commands more efficiently to avoid wasting time.
We do a second test after spending many months on development and design. What do the users think of it now? Far too complicated! Planning ahead wasn’t that important after all.
So we scrap all the new functionality we’ve designed and start over.
In other words
Interviewing is a specialist skill. People don’t know what they need, but they’ll tell you fascinating stories. So as a researcher, you have to understand how to uncover reliable information.
In a design team, you need people who understand the nature of research. Without that knowledge, you’ll be basing your design decisions on non-information. In that case, you’re better off – and you’ll save yourself a lot of money – by not doing any research at all.
And what if we don’t have a budget to hire a specialist for every decision we make?
Then you look for smart ways to perform meaningful research with limited resources:
- You appoint an experienced researcher to act as a coach to inexperienced interviewers
- You get an experienced researcher to design a standard research protocol
- You send the most motivated employees on an interviewing course
- You organise regular meetings with the employees who do research so that they can learn from each other
- You invite experts to conduct open lectures during office hours
- You record interviews on film and analyse them with your team. What went well? What could be improved?