There was a time when user testing wasn’t particularly practised. There was a general view—sometimes not explicitly stated—that we could devise a product or interface within our offices, launch it, and it would be good. Fortunately that’s largely changed. There has been a real shift in recent years from thinking we know the answers to testing hypotheses with real people.

But sometimes we go too far, and end up at the other extreme. I sometimes come across people and teams who will not make any user experience decision until it’s been tested with users. They have gone from “we know everything” at one end of the scale to “we know nothing” at the other.

Not only is this silly, it’s damaging to the organisation, the product and the users.

When we build or adjust a feature we’re making a bet on how it will be received. If we’re not sure then we should test it; we should check our assumptions, too. But it can often make sense to build and release it without user testing. This is because user testing takes time; the more time we spend on that the more the release is delayed, and so user benefit is also delayed. The balance is not always on the side of user testing. This is especially true when we see poor user experiences.

If user testing of a feature is going to take four weeks then that’s four weeks less time our users will have with the feature, and four weeks of us losing the benefit, too—of user engagement, increased sales, or whatever. There is an opportunity cost, too. If our user researchers are working on this then there’s something else that they’re not working on.

Many in our teams have expertise in our users and their experiences; we can often trust them to make a judgement. We can often trust our own experiences, too. We should be wary if we think one form of presentation is more effective than another—that’s probably worth testing. But we can usually be confident when we sense something is a bad experience. If we’re 95% sure a proposed widget, say, is clunky, let’s seek to improve it immediately rather than waiting for external users to tell us what we privately knew anyway.

The growth of user testing has been hugely positive for product organisations and users. But as with so many activities its cost is greater than zero, and we should not use it to pretend we are empty vessels, ignorant of all digital experiences.

Photo by cammon