I’ve handed Google 30% on every sale for fifteen years This month it drops to 20%, and I have complicated feelings.

The mobile tax everyone complains about is finally moving. Once I actually ran my numbers, finally came with a lot of small print.

For about as long as I’ve shipped apps, the deal has been the same. You make something, someone pays for it, and 30% of that vanishes into the store before you ever see a cent.

You stop noticing it after a while, the way you stop noticing the rent line on your bank statement. It’s just the weather. The cost of standing in the one place the customers actually are.

That’s changing this month, on Android at least. Google’s dropping the Play Store commission from 30% to 20% across the US, the UK and the EEA by June 30. New installs through their App Experience and updated Play Games programs can go as low as 15%. Subscriptions drop to 10%.

And the big one for me, you can now run your own billing right alongside Google’s, with Google taking a 5% cut for the privilege instead of the full freight.

I have wanted this for over a decade. So why did I close the announcement feeling oddly flat? Because then I opened a spreadsheet.

The bring-your-own-billing math is closer than the headline

Here’s the daydream every app developer has had at some point. What if I just took the payments myself and kept the whole thing?

So let’s actually do it. If I bring my own billing, Google still wants 5%. Then my payment processor, Stripe say, takes roughly 3%. That’s about 8% before I’ve handled a single edge case.

Set that against Google’s own billing at 20%, or 15%, or 10% on subscriptions, depending which bucket I land in.

For a one-time purchase paying the full 20%, doing my own billing wins, clearly, even after the 8%. For a subscription app already getting the 10% rate, the gap basically closes, and that’s before I remember everything Google’s cut was quietly paying for.

Sales tax and VAT in every jurisdiction, calculated and handed to the right government. Chargebacks. Fraud. Refunds. The furious email when someone’s card gets declined in a country I’ve never set foot in.

Google did all of that, clumsily sometimes, but I never had to think about it. Take the billing in-house and every bit of it is suddenly my problem, and my problem has a salary attached that never shows up in that tidy 8%.

So the honest answer to the daydream is: it depends, and for a lot of apps the new in-house option is a near-tie that hands you control and a support burden in roughly equal measure. Which is a far less satisfying sentence than abolish the tax.

The small print is doing a lot of work

The number that gets the headline is 20%, but look at how many numbers there actually are.

Thirty drops to twenty, or to fifteen for new installs if you’re in the right program, or to ten for subscriptions, or to five if you bring your own billing and still pay your processor on top.

The fee is now a small menu, and which item you’re served depends on what kind of app you have, whether the install is new or existing, and which program you remembered to sign up for.

It’s also staggered by geography. The US, UK and EEA get it by June 30. Australia by the end of September.

Korea and Japan by the end of the year. The actual whole world, Google says, by September 2027. So depending on where your users live, the tax dropped might be a 2027 sentence rather than a this-month one.

All of this traces back to Google settling with Epic late last year, and to be fair to Google, they’re rolling it out in public instead of dragging their heels waiting on every last signature.

There’s also a separate seven-hundred-million-dollar settlement with a stack of state attorneys general that got approved in the spring. The pressure was real and it worked. I just want to be precise that what happened here is a renegotiation, not a liberation.

The part nobody on Android wants to say out loud

The other half of my revenue comes from the App Store, and Apple has done approximately none of this.

Apple is still fighting it, still litigating the external-purchase question, still running the model where the cut is the cut.

So as a cross-platform developer I get to enjoy this win on exactly one of the two platforms I ship to, which means my actual blended take-home moves a lot less than the Android headline makes it sound. The 30% tax cracked on Android. On iOS it’s still mostly load-bearing, and iOS is where a lot of the money lives.

The genuinely interesting long game isn’t the percentage anyway. It’s the Registered App Stores piece, third parties being allowed to run real stores on Android with a proper install flow.

If that matures, the leverage shifts in a way a few points of commission never will, because then the store has to actually compete to keep me. That’s the move I’m watching, and it’s further out and blurrier than the fee cut.

What I’m actually doing about it

Re-running the pricing, for real this time, not as a daydream. Because the right answer genuinely changed this month, and I’d be leaving money on the table to keep operating on 2024’s assumptions.

For my subscription app I’m probably staying on Google’s billing at the new 10%, because once I price in the tax-and-fraud headache, paying Google a dime on the dollar to make all of that vanish is, annoyingly, a decent deal.

For the one-time-purchase side I’m seriously costing out my own billing, because 20% versus 8% is real money at any kind of volume. And I’m putting the per-country rollout dates in my calendar, because my users aren’t all sitting in the lucky June-30 zone.

Anyway. Fifteen years of watching 30% come off the top, and the day it finally moves I spend the afternoon in a spreadsheet feeling like I should be happier than I am. The tax went down.

I’ll take it. I just learned, one more time, that good news and simple almost never ship in the same release.