Tristan Corriveau is a Maine-based entrepreneur who has worked for local startups, volunteered with the Startup Maine conference, and founded The One Gallon Soap Company. His latest project is Klapa, a startup for group payment splitting.
Klapa is a tool to give people making large purchases as a group an easier method for checking out of online businesses. I was excited when Vinko told me he was interested in coming on board with the project. However, before we built anything, he had one condition: we had to test our ideas with prospective users. So when we were done filling out our business canvas, we decided to set up at our local shopping mall and get real people’s perspectives. If Vinko hadn’t gotten us to take a more scientific approach, then things could have gone in a completely different direction from day one. We might have ended up with a beautiful product design that no one knew how to use!
When I started The One Gallon Soap Company I didn’t think much about testing because I thought people had pretty set expectations of what liquid hand soap should look and feel like. There were also so many variables: container design, color, viscosity— how would I even choose what to test? Even if I did design a good test, I knew it would be a lot of work to repeat it enough to be sure of the results. One of my favorite books on behavioral economics is called the Undoing Project. It includes this great description of something called the law of small numbers, a “human glitch” where we draw the wrong conclusions from limited information. Without organized testing, it’s so much easier to be misled by this glitch. Anecdotes and unquestioned assumptions are incredibly risky evidence to build a product on, but too often the risk gets overshadowed by fear of being wrong. Embracing systematic testing with Klapa helped me get past these glitches and see that this fear is misplaced: the earlier and more often you manage to prove your assumptions wrong, the stronger your product gets.
First, it’s important to identify every factor in your test that influences the participants. For instance, when we were finding people to survey at the mall we would ask them, “have you ever bought something online where you had to front the money for a group?”. We didn’t realize it at first, but this question was psychologically priming our subjects. It was only when we tested our UI with a less specific opening question that we found out people couldn’t make it through on their own. It was pretty upsetting to find out how easily we had fooled ourselves with a design that didn’t work, but after the sting of being wrong wore off, I saw how valuable our first negative feedback was. It turned out we had placed the group purchase option so it was competing with the “checkout” button, and we were able to fix something that might have hurt our user experience a lot.
As you start to realize how useful negative feedback is, you’ll discover different ways to tweak your interviews to avoid discouraging honest feedback from your users. When we were talking to people, I had to overcome a strong impulse to help them understand our design. If someone asks you why a button is there, it’s so tempting to explain it, but you’ll get much more useful information if you just throw the question right back at them: “why do you think it’s there?” You don’t just want interviewees to reflect your knowledge, you want them to contribute theirs. Even knowing this, it’s so easy to coach people through your designs via unconscious verbal and nonverbal cues. I found that testing with a teammate was a great way to spot these cues and keep each other in check and avoid biasing our interviews by being too helpful.
One of the hardest things to master about user interviews is paying attention to what people do, not just what they say. Again, this is where having a partner for testing really comes in handy. While I was busy conducting the interviews and taking notes, Vinko was able to notice all these nonverbal cues. Sometimes you’ll have a successful test but you can tell the user isn’t really excited about it. Our first UI prototypes were just static images, and even though we could simulate the flow for people, they seemed really passive. Without recording the nonverbal and emotional details, the feedback would have looked way better than it actually was. Because we could see that the nonverbal feedback was negative, we decided to code up a minimal interactive version of the prototype and people were a lot more excited and really engaged with it since they could click through the menus themselves.
When you’re user testing, you’ll have these moments over and over again when you realize you’ve narrowly avoided taking a wrong path. I can’t stress enough how helpful it is to talk to real people at the beginning of a project, and how much it helps to have teammates to check your methods and make your tests more rigorous. If you get negative feedback late in your development process it can be really costly to go back to the drawing board, but if you're just starting out you’ve got nothing to lose and everything to gain. Even tried and true products like hand soap can be improved with testing, but for a first-of-its-kind product, it’s absolutely essential to make a research a habit. Don’t be surprised if you find yourself actually getting excited about proving yourself wrong— it just means you’re on your way to the innovation mindset!