I list six mistakes of mine together with ideas on how to avoid them. These ideas are bound by one key principle - Reduce the effort for the reader.

ChaGPT is indeed pretty good at coding simple stuff. It helped me build a tool to solve a tiny inconvenience I experienced as a language learner. Sharing my process of building the tool in this article.

In hindsight, it all seems quite obvious: allowing users to do whatever they want with your platform (or parts of it) is not a good idea. Still, in the heat of the moment, you might forget this, and the idea of giving users as much value as possible, as soon as possible, completely takes over.

Sometimes I wish that after releasing a feature (and especially something that changes the current product in any way), you can tune out everything except bug reports. The problem it would solve is the most vocal users fallacy - when you only hear the bad things about the new update you just shipped.

I've recently had a small but remarkable mental breakthrough about the nature of features we build in B2B SaaS. Unlike B2C, it is not only about constant optimization and testing as many high-quality hypotheses as you can (who would've thought). Instead, we're looking for the balance between what sells and what retains.

When - exactly after I've asked my question. How - by listening to myself, simply reading what I have in my script, and embracing the awkward silence. Why - to make sure I don't waste that precious user interview, and increase the chances of getting a valuable insight.

Having a list of keywords you react to helps navigate the conversation and extract insights. Going against your reflexes by saving follow-up questions for later, admitting your mistakes, and muting yourself make the interview run smoother and let you control it.

I have built this website thanks to a great YouTube tutorial, the help of developer friends, and knowing exactly what I need to learn next.

My friend and I didn't have an idea, but we really wanted to try starting a business together. The original plan consisted of interviewing a few people inside a segment, coming up with an idea, testing it using the cheapest ways we could find, and making an estimated guess whether the business idea has a future or not.

How different is an action that a respondent is being asked to do from an action that a respondent would do in real life? Perhaps, user researchers should take their UX testing findings with a grain of salt.