There's a more general takeaway, and it's one every developer discovers for themselves, sooner or later: - People don't read what's in front of them. I've see

There's a more general takeaway, and it's one every developer discovers for themselves, sooner or later: - People don't read what's in front of them. I've seen this emerge in a vast array of fields. No matter how much we highlight specific details, for all our efforts in red-flagging irreversible actions, folks will often blitz past a confirmation dialog, nag screen, or notification message, without internalising the details or the risks. For those in financial technology, as in this specific example, irreversible actions also extend the attack surface for fraud. Even the brightest minds can be lazy (some might even say it's a feature, not a bug) and one should never rely upon the opposite. We consequently face a design choice, for all irreversible (or hard-to-reverse) actions, the most common options being: a) allow a grace period; b) redesign, if possible, to make it user-reversible; c) build a forcing function for diligence[1]; or d) expect support tickets about that feature. The default is (d), and the helpdesk won't thank us, since the workload generally scales linearly with growth at a high opportunity cost. [1] e.g. https://en.wikipedia.org/wiki/Two-man_rule https://en.wikipedia.org/wiki/Two-man_rule