Policy Harm
Much of my personal frustration with the current state of web ideas stems from the lack of awareness around policy harms. The companion to policy harm is administrative burden. I’ve written a lot about friction, and it's part of these ideas too. Namely: if you can frustrate someone enough, it’s easy to get them to do what you want.
In public policy contexts, burdening people or forcing them to have “skin in the game” gives bureaucrats the sense that they’re weeding out grifters. It’s the freeloader problem applied to governance. Outside of policy, policy harm is still present—especially in the era of “move fast and break things.”
What is the cost of policy harm?
What we have to reckon with as designers is: what are the implications of policy harm, and who pays the cost?
Usually, users are the ones who have to deal with the unintentional friction built into experiences. But bounce rates or Reddit complaints aren’t a meaningful way to quantify the cost of these failures.
A few years ago, I worked on an onboarding experience for single sign-on (SSO) where ~25% of users quit the flow before completing it. Because the user base was in the millions, that loss was considered “acceptable.” Later, the team improved the flow—and exit rates dropped significantly.
Now shift from digital to real-world experiences augmented by bad UX. Think about navigating mass transit in an unfamiliar city. If you’ve ever taken a train abroad, you know the onboarding experience is