This week, I visited a startup whose office had a very unusual feature: a usability lab. There I was in a soundproofed room with a table in the center flanked by two chairs, one for a user and one product manager or UX researcher. On the table, a constellation of web cams record a user’s facial expressions and interactions with a mobile phone or laptop while a microphone captures the user’s voice.

It had been years since I had been in a similar room at Google, where usability studies are more or less required for major product releases. We would invite five users at a time to visit us periodically over the development of the product.

We asked each visitor in turn to sit down and begin using the new product or feature all the while speaking their thoughts. At the beginning and the end of the project, we wouldn’t intervene, but watch the user’s first experience in silence with great interest. Other times, we focused the user on a particular element of the product by asking them to perform a task or asked pointed questions about the design of an input table or a workflow.

The results of the studies always confounded me. Users would stumble through “simple” parts of the product and point out weirdnesses we thought would be accepted as universal design affordances. But that was the point: to invalidate our assumptions and find the points of friction in the product.

Stepping out of the startup’s usability lab, I left wondering why labs aren’t more common in post-Series A companies and incubators - and why I hadn’t suggested the idea to the entrepreneurs I work with. Despite the work required to find and study users during our product work at Google, the effort always yielded a handful of key insights we might not have discovered otherwise.

It is a stroke of genius on the part of this startup to dedicate a part of their office to the study of the use of their product in users' hands. After all, that is the most important part of product design: understanding the user.