Independent UX consultant Ida Aalen recently co-founded a video conferencing startup. At Pixel Pioneers Bristol she will explain how to do user testing with limited resources and how to get the whole team involved. We caught up with her to find out more
How did you come up with the idea for Confrere and what does your role involve? The idea comes from our CEO Svein Willassen. After four years as founder and head of appear.in, he realised that video conferencing systems were all made for internal team communication. There were no good solution for professionals meeting their clients, for example doctors, bank advisors or tutors. Last summer, Dag-Inge Aas, former tech lead of appear.in and I joined him as co-founders, and in July 2017 we founded Confrere.
Since we’re a small team, I get to do a lot of stuff — but a lot of stuff that I love. I’m heading the branding and design work, I’m making sure our product is user-friendly and accessible, and I’m also working on our external communication, for example in social media.
How’s the product coming along, and what are some of the lessons you’ve learned so far? Above and beyond expectations! We released the beta in November, and we already have paying customers, like doctors, psychologists and advisors within different fields.
One really import lesson is how essential it is to test in the field. We were thinking that the product we had would be perfect for doctors. But when we went to their offices, we realised so many of them of course didn’t have webcams or microphones. Or were using old browsers. And most of them have a speciality web connection where it turned out we had to be whitelisted. And we found a ton of bugs. We’d never realised any of this if we’d just stayed at Startuplab, where we’re based.
The other thing we’ve realised is that this isn’t really a technical problem, nor a design or usability problem. The most important thing is to build a tool that fits neatly into people’s existing workflow. For instance, for a doctor’s office that mainly has face-to-face consultations, how do we make sure adding video consultations to their services doesn’t just mean more hassle for the doctors and their assistants?
How can designers convince their boss that user testing is necessary? First of all, having experts in your team doesn’t mean they’ll find all the usability issues. Comparisons have shown that a third of the problems the experts finds are actually false alarms, and even worse, that half of the usability issues found in user testing were not identified by experts.
So you’re simply taking a big risk when you launch something without testing it. There might be serious issues you haven’t identified.
In addition, user testing is a lot easier, less time-consuming and cheaper than people think. You can learn a lot just from a couple of hours. Learnings that can save you both money and worries later on.
What are your top three testing tools and why? My favourite is simply to jump on a bike and go where my users are and meet them in their natural environment. You should always include some form of user testing that means actually interacting, meeting and talking to your users.
If you’ve ever fought about how to name a menu point, or where in the site architecture a certain page should be placed, you’ll love Treejack surveys. With this tool, you recreate the information architecture within the survey and give users a task to solve, e.g. “Where would you click to find career opportunities?”. Afterwards you get a visualisation of where people looked in the information architecture.
Another tool is first click-tests like Chalkmark. You show people a picture, be it a screenshot or a drawing or a mock-up, and you give them a task, e.g. “Where would you click to find the pricing?”. Afterwards, you get a heatmap of where people clicked.
How do you get the questions (and the set-up) right without leading the test subject and bringing in your own bias? My suggestion is to first write down all the thing you want to find out. A typical mistake is to ask the user a question that is too close to what you want to know: “Can you put this item in the shopping cart?” or “Can you find the logout button?”. But after you’ve written down what you want to find out, you should then think about how to frame the question. Give the user a scenario to solve, rather than asking them to find something. So we could for instance ask “What would you do if you liked this item and you decide you want it?” or “What would you do after you’ve finished?”.
How do you analyse the data from user testing sessions to ensure you can actually use it to improve your product? The analysis should happen as soon as possible after the user test. I try to do it the same day. Two days later, you’ve already forgotten a lot. After the problem is clearly formulated, assign someone to come up with a solution to the problem. Too often we jump to conclusions about solutions when we observe a user test. But the user test is about finding problems, not solutions. Don’t close that window too soon. It’s also a lot more engaging for designer and developers to get a problem to work on, instead of just being told to implement some ready-made solution.
What’s the best piece of advice you’ve ever received? I think it’s quite common in the beginning to feel that a user test that results in you finding a lot of usability issues is a failure. But this just means you’re framing it wrong, you’re thinking “I made something crappy”. It’s actually the other way around. A user test with lots of issues is a great success: you found all of these problems, so that we can fix them! And you didn’t make something crappy, you have a process that works, and you’ll make something you actually know works. I’m always really happy when I find usability issues, even when I’ve helped build the thing we’re testing.
What can people expect to take away from your talk at Pixel Pioneers Bristol? My goal is to give people practical, hands-on tools, tips and advice. I want as many attendees as possible to feel they’re already ready to do more testing when they get back to work on Monday. Let’s see if I succeed!
Pixel Pioneers Bristol on 8 June features 8 talks, preceded by a day of workshops with Michael Flarup (Designing Better App Icons) and Joe Leech (Psychology for UX and Product Design). Ticket bundles, group and student discounts are available.