The real cost of running an indie app nobody warns you about

Here's the honest version of indie app running costs: the fixed ones are boring and small — a few store fees, some tooling, a domain. The scary one is variable. Every saved video needs heavy processing, and that bill grows with every user. That single line is what decides whether an app survives. The rest is rounding error.

How much does an app cost to run, really?

When people ask how much an app costs to run, they picture one number. There isn't one. There are two buckets, and they behave completely differently. Fixed costs you pay whether you have ten users or ten thousand. Variable costs climb with every single thing your users do. Mix them up and you'll budget for the wrong disaster.

We're Mrinmoyee & Sudipta, the two-person team behind Trott. We built it in public, shipped it to both stores, and watched the real numbers land. So here's the breakdown nobody hands you when you start.

The fixed costs (the easy part)

These are the app development expenses everyone warns you about, which is exactly why they're not the problem. You can write them on a napkin:

• Apple Developer Program. An annual fee to ship on the App Store. A known, fixed line item.
• Google Play. A small one-time registration fee to publish on Android. Pay once, done.
• Subscription tooling. Things like RevenueCat for handling purchases — free until you're earning real money, then a small cut. Fair, and it scales with success, not with risk.
• Domain and website hosting. A domain renews for a few dollars a year. A simple marketing site hosts for free or close to it.

Add it up and the fixed side is a small monthly fee plus a couple of annual ones. You could cover it with a part-time weekend. Nobody's app dies here. We've never once stared at the Apple invoice in fear.

The variable cost that actually keeps you up

Now the one nobody warns you about. Trott reads every video you save — the frames, the spoken words, the caption — and writes down what's actually in it so you can find it later. That reading is heavy work. It's compute, and it costs money. Every time. Then the result has to live somewhere, so there's storage on top, and storage only ever grows.

That's the trap. A fixed cost is a flat line. This one is a slope. Ten users is nothing. A thousand users each saving a few videos a week is a real bill that arrives every month and keeps climbing. The work is genuinely useful, which is the whole point — but useful isn't free, and the meter runs on every save whether that user ever pays you a cent.

The harsh math: cost per user

This is where it gets uncomfortable. Once you know roughly what processing one user's saves costs you per month, everything else is arithmetic. You're not asking "can I afford to keep it running." You're asking "how many of these people need to pay me before the math works."

Run it honestly and the picture is stark. If most users never subscribe — and on a free-to-start app, most won't — then a small slice of paying users has to carry the processing bill for everyone else. Get the cost-per-user wrong by a few cents and your break-even point doubles. We learned to write this down before adding features, not after. "It works" and "it works without losing money on every video" are two different milestones. It's also why we tied how we price Trott straight to this number.

What you do about it

You can't make the heavy work free. You can make sure you only pay for it when it earns its keep. The levers, stated plainly:

• Don't pay for the same work twice. If two people save the same viral video, you shouldn't pay to read it twice. Do the expensive work once, reuse the result.
• Only do the heavy work when it's needed. Early on we did everything, eagerly, the moment a save arrived. Some of it nobody ever looked at. Doing it only when someone actually needs it cut waste we didn't know we had.
• Be patient about it. Spread the work out instead of treating every save like an emergency. Patient work is cheaper work.

None of these are clever. They're just the difference between a cost-to-run-an-app curve that bankrupts you and one that bends. We didn't get them right at first. We got them right after watching the bill.

If you're about to build something

Budget the fixed costs in an afternoon and then forget them. Spend your real worry on the one line that scales with your users, because that's the line that decides whether you have a business or an expensive hobby. Know your cost per user before you chase your thousandth user.

That's the part nobody warns you about. Now someone has. If you want to see where all that processing actually goes, give Trott a try and save a reel — that's the whole cost story in one tap.