Should I build an Android version?

Appray is iOS and iPadOS only. It was always going to be, mostly because I know Swift and I do not know Kotlin, and learning a second native platform while also running support, writing this blog and doing everything else a one person company does felt like the kind of ambition that gets a founder nowhere in particular. That reasoning held up fine for the first few months. It is starting to feel less like a reason and more like a habit I have stopped examining, which is exactly the kind of thing I built this app to catch other people doing.

The maths everyone will tell you, and it is mostly right

Ask anyone who has shipped on both platforms and you get roughly the same answer. Paid iOS apps in a niche like mine tend to out earn their Android equivalent by a wide margin, sometimes by five or ten times, because the App Store audience is smaller but pays for things, and the Play Store audience is larger but expects free with ads. Appray is a one time purchase with no ads and no subscription. Put that model in front of an Android audience and the honest expectation is a fraction of what it does on iOS, for the same amount of app.

I do not think that maths is wrong. I have read enough interviews with developers who ported a paid app over and watched it earn five percent of the iOS number to believe it. What that maths does not tell you is whether the decision in front of you is actually about money, and mine is not, not primarily. Nineteen people did not email me about revenue. They emailed me because they use Android and cannot install the app, which is a different problem wearing the same clothes.

What the request is actually a proxy for

This is the same question I tell other founders to ask about any feature request, and I nearly skipped asking it of myself because "build an Android app" sounds too big to be a proxy for anything smaller. It is not too big. I read through all nineteen emails properly instead of skimming for the word Android, and they split into two piles that I had been treating as one.

Twelve were from people who do not own an iPhone and simply cannot use the app, full stop. There is no smaller fix for that. Either Appray runs on their phone or it does not, and no clever workaround changes that.

Seven were from people who own an iPhone for work and an Android phone personally, or switched phones recently, and wanted their Appray data to follow them. That is not actually a request for an Android app. It is a request to not lose eight rounds of answers and a Build, Kill or Prove verdict because of which pocket their phone happens to be in. A plain export file, something you could open on any device and hand to a new install later, solves that for a fraction of the cost of a second native app, and I had never separated it from the Android requests because both arrive with the word Android in the first sentence.

What building it would actually cost, said plainly

I am not going to pretend a rewrite is a vague inconvenience. Appray has no server and no account system by design, everything runs on device, which is the whole privacy pitch. Porting that to Android with Kotlin and Jetpack Compose is not translating files, it is rebuilding the interview flow, the verdict logic and the on device storage a second time, in a language I do not currently know, while learning a platform I have never shipped on. A realistic estimate for me, alone, evenings and weekends around existing support and marketing work, is four to six months before there is anything I would be comfortable putting in front of a stranger. That is not a guess dressed up as a number. It is roughly what the original iOS build took me, and I would be doing it again with none of the Swift experience carrying over.

Four to six months is also, not coincidentally, close to how long it took me to go from an empty project to an app that thirty one features long list I wrote about a few weeks ago. I would be spending a second MVP's worth of time serving an audience I have not yet proven wants the paid, no AI, no account version of this app at all.

The test I actually ran

I used the same Prove step I write about for everyone else's features, on my own company, which felt more uncomfortable than I expected it to. A landing page with one paragraph explaining what Appray does, a line stating clearly that it does not exist for Android yet, and an email field for anyone who wants to know the moment it does.

I have run it for three weeks. Forty one signups, against nineteen emails before the page existed and roughly four thousand App Store page views in the same window. That is a real number and a small one, and it tells me something neither pure enthusiasm nor pure dismissal would have. There is a genuine audience. It is not currently large enough to justify four to six months of a second native build for an app that has not yet proven it can support the founder building it full time on one platform.

Where I actually landed

Not Android, not yet, and I am done pretending the reason is purely technical. The export file for the seven people switching phones is going in the next update regardless of what happens with Android, because it is cheap, it solves a real and specific problem, and it does not touch the no account, nothing leaves your device promise that is the entire point of the app.

The bigger build stays on the landing page, genuinely watched rather than quietly ignored, with a number I check instead of a vague sense that "people keep asking." If it crosses something like two hundred signups, or if Appray on iOS alone gets to a place where four to six months away from support and marketing is survivable rather than reckless, the answer changes. Until then it sits in the parked list from that MVP post, with one line about who asked and why, instead of quietly rotting as a Kill I never admitted to making.

The part I got wrong for months was not the no. The no might still be right. It was collapsing two different requests into one big scary project and refusing to look closely enough to notice that most of them were asking for something I could actually give them next month.