Spicy take on UXR and Teenage Engineering

At a recent talk at Config, the Teenage Engineering CEO had some takes on building great products that didn’t go down well with some folks (and were loved by others).

Specifically, he said he doesn’t do User Testing.

A few thoughts:

1️⃣ The second I heard him say it, I knew it would be an issue. User Testing has become sacred in the startup world. Something that is unquestionable. Especially by designers. Personally, whether right or wrong, I LOVE that there are different takes on building products. The whole point of someone like him being on stage is to challenge our practices. You don’t have to love it. But learn from it.

2️⃣ There’s a HUGE distinction between not doing traditional user research and not knowing your users. If you think for a second that Jesper doesn’t know his users, you’re wildly misled. He went on to share how he sometimes spends YEARS getting into the weeds of a product and what it takes to use it. This is not flippant, “I build whatever I want” mentality, but a deep understanding of the user problem.

3️⃣ CEOs are constantly getting a litany of feedback and opinions. They get it from users, from partners, from customer support, and every other function. In my experience, many founders are wildly well informed about their users and their needs. Sometimes they don’t use the same language as researchers and don’t speak from user’s perspectives. Instead, they often act by “gut”. This gut is not just a feeling though. It is a gut informed by living and breathing the market and its users every single day.

4️⃣ TBH, Designers are often not on the front lines and not naturally speaking to users on a daily basis. We forget this. So for us, we have to construct discreet conversations with users. That’s great. And the rigor is hugely valuable. But it is by no means the only way to understand the people that we’re building for.

5️⃣ But people LOVE Teenage Engineering products. And by the way, Steve Jobs was known for making similar statements. That’s the amazing piece for me. Why? Not because they didn’t know their users. But because they knew them intimately and imagined an alternate reality. They take risks that their users wouldn’t take, or even suggest. They see a technological future and bring people along for the ride.

6️⃣ Teenage engineering can also take this approach precisely because they are not trying to build for everyone. We’ve become so hyper focused on HUGE scale, and the idea that success is equal to global acceptance. We’re obsessed with growth. They seem to be taking a perspective that they are more focused on quality. Some have even described it as a fine art piece. Let it be what it is. Let’s all accept more people’s paths. Let’s celebrate that people are able to pursue something with passion. THAT seems to be where great products come from.

7️⃣ If you take VC money, the natural pressure is for global scale. The biggest TAM possible. If that’s the goal, then building products that are infinitely scalable, and usable by everyone becomes the primary focus. You can’t possibly know what 100MM people want, so you talk to as many as you can, and generalize their feedback. My hot take is that has led us to more and more generic products that are less and less enjoyable for most people. Not objectionable. Very usable. Just not very delightful.

So what’s my take?

👉 Personally, as with most things, I think there’s a balance. I’m a huge fan of talking to users, understanding their problems, and building excellent solutions. But this perspective of niching down and building something truly enjoyable for a smaller group of people can also sometimes have outsized impact. MANY products have been built this way.

👉 Passion is a great driver. If you LOVE the product you’re building, it often leads to a very good product, especially if you are tuned in and not designing in a vacuum.

👉 We should be open to products having strong opinions. To reflecting the people that built them. This will mean they may not be suitable for every user, and that has real downsides. But it’s definitely ONE way to build software.