Predictable behavior

One of the things UX/product designers do is predict the way people will use things. It helps us prevent errors and usability problems. It’s one of the core skills of a good designer, and one of the problems we must solve first; before we decide what it looks like, or how it will make you feel. I mean…unless you’re a bad designer. Then you design things for how they will look in the pages of your favorite design magazine (if those are still even a thing), or to placate your most opinionated stakeholder. I guess it depends on whose validation you crave the most.

Anyway. I digress.

We should be designing things based on how we expect they will be used. It seems simple enough, but it’s this very basic idea at which we fail again and again, especially when we design online systems. We consistently fail to predict, and design for, people using the systems we design inappropriately.

It occurred to me this morning, as I was reading yet another exposé detailing Facebook putting its thumb on the scale in order to avoid receiving criticism from right-wing media pundits while listening to a podcast wax comically about the nudity police on Instagram; that this failure is not a failure of design to predict the behavior, but a failure to accurately calibrate the idea of “appropriate.”

Or, said another way; the problem is not that we can’t predict the behavior and design for it, it’s that we don’t agree that the behavior is inappropriate.

In this example, we seem to all be aligned on the idea that online services that allow children to have accounts and post content should not allow nudity, but we’re not aligned on whether or not we should allow that same platform to be used to spread hate speech, organize domestic terrorism, or manipulate people’s emotional health.

When you think about it, especially considering the problems that a platform chooses to fix, versus the ones it chooses to fix (or fix poorly); what they are really telling you is which uses of their platform are appropriate or inappropriate. Or, at least they’re telling you which behaviors they’ve decided aren’t deal-breakers because they’re important to their most opinionated stakeholders.