People asked for a Face ID lock. That wasn't the actual problem.

Appray holds whatever you're working on, project names, the answers you gave in the interview, the verdict it handed back. For some people that's a hobby idea. For a few it's the next thing they're about to quit a job for, written down before they've told anyone. A Face ID lock is a reasonable thing to want on that, and eleven separate people asking for it over four months is more consistent agreement than most feature requests get from me.

I had the word "lock" in my head and went straight to the obvious design. iOS has a LocalAuthentication framework built for exactly this, Face ID or your passcode as a fallback, gate the app behind it the moment it opens. I'd got as far as sketching where the prompt would sit before I did the thing I tell other people to do and actually went back through all eleven emails properly instead of skimming them for the word I already expected to find.

None of them described a stolen phone

I'd assumed the scenario was the obvious one: phone lost or stolen, stranger gets in, reads your business idea. But a lost phone is already locked by the device passcode before Appray ever enters the picture, and if someone's device passcode is compromised, an extra Face ID prompt on one app among forty isn't doing much additional work.

What eight of the eleven emails actually described, once I read past the word "lock," was handing an already unlocked phone to someone else for a few seconds. Showing a cofounder a photo and watching them swipe one screen too far. A kid grabbing the phone to play a game and landing on whatever was last open. Demoing something unrelated at a meetup with Appray sitting in the app switcher from twenty minutes earlier, project name visible in the little preview thumbnail iOS shows you mid-swipe.

That is a completely different problem from the one I'd started designing for, and it was sitting in my own inbox in plain English the whole time. I'd just read "lock" and gone looking for the mechanism rather than the moment.

Face ID would have made this worse, not better

Here's the part that stopped me cold. A Face ID gate on app open solves the stolen-phone case I'd imagined and does nothing for the handed-over-phone case everyone was actually describing, because in that scenario the phone is already open to Appray, or one swipe away from it, in someone else's hand. Nobody re-locks an app mid-handover to ask their four year old to re-authenticate before showing them a cartoon.

Worse, a prompt on every single open is friction every person pays on every ordinary use to guard against a ten-second scenario that happens to a handful of people a handful of times a year. I'd have shipped something that annoyed everyone daily and protected nobody at the exact moment they needed it.

What actually fixes a two-second glance

The thing iOS shows in the app switcher is a screenshot it takes automatically the instant you leave an app. Banking apps have dealt with this for years by swapping in a plain screen the moment the app resigns active, so the screenshot the OS grabs shows that instead of your balance. I did the same thing: the moment Appray goes to the background, it swaps the real screen for the logo on a plain background, then swaps back the instant you return. The project list and every project name in it are gone from the preview entirely, with nothing to unlock and nothing to remember a passcode for.

It took an afternoon, most of it spent making the swap happen early enough that the screenshot never catches the real screen mid-transition. No LocalAuthentication, no new permission dialog, no daily friction. It covers the scenario eight of eleven people described and does nothing for the scenario the other three mentioned explicitly, which was closer to wanting a diary-style lock out of habit rather than a specific worry, and I told them honestly that I'd solved for the common case and left the rest open.

The actual lesson

The word someone uses to ask for a feature is their best guess at a mechanism, not a description of the problem. "Lock it" is a solution already, borrowed from every other app that offers one, and it arrives in your inbox sounding so complete that rereading it feels unnecessary. I nearly built the thing eleven people asked for by name and it would have fixed almost none of what they were actually worried about.

The question that would have caught this earlier than it did is the one I now ask myself before opening Xcode on anything requested by name: what were you doing, physically, in the moment this mattered to you. Not what do you want. What were you doing. Four months is a long time to go without asking that.