SquareCat
Miles Corran · 
“How much will it cost?” is the first question most people ask about an app, and the honest answer is that it depends on decisions you haven’t made yet. Two education apps that look alike in a sentence can differ a lot in the work behind them.
This post doesn’t give prices. It walks through the scope decisions that move the cost most, so you can see which ones matter for your project and where you have room to start smaller.
Platforms
Web, Android, iPhone or all three is usually the biggest single decision. Each platform has its own screens to test, its own store rules and its own release process.
Sharing code helps. Our apps run on the web and Android, and the Android apps are built from one shared mobile codebase. When we added CertBrain on Google Play, the third app from that codebase was mostly configuration. That’s a saving that comes from earlier investment, though, not one a first app gets for free.
A common way to start smaller: launch on the web, where a change ships without a store review, and add mobile once you know what learners use.
Content volume, and how it gets in
The amount of content matters less than its condition. A course in a consistent structured format can be imported. A course spread across slides, PDFs and documents in different styles has to be cleaned up first, either by you or by us.
For Finance Brain, we wrote an importer that turns a structured content library into courses, lessons and multiple-choice decks, images included, and re-imports when the source is corrected. An importer is work up front that pays back on every later update. Getting your content into one consistent structure before the build starts is one of the most effective ways to keep this part small.
Accounts
An app with no sign-in is simpler. An app with accounts needs sign-in, recovery, privacy terms, a way to delete an account, and progress that follows a learner between devices.
Accounts are usually worth it for a learning app, because progress is the point. But not every feature needs one. CertBrain’s diagnostic quiz needs no account and stores nothing: the browser scores it. Deciding which parts work signed out is a scope decision in its own right.
Payments
Taking payment adds subscriptions or purchases, receipts, what happens when a payment fails, and the rules about what’s free and what’s locked. Mobile stores have their own payment rules, which adds more again.
You don’t have to build it all at once. CertBrain’s premium content was gated from the start by a single check, with nobody holding premium until billing existed. The paywall worked before payments did, so billing could follow later without changing the content.
AI features
AI features are cheap to prototype and easy to leave open-ended. The running cost grows with use, so the scope question is how you cap it. In Braineos, AI features are part of the paid plan and limited by a monthly quota. We wrote up the patterns in adding AI features without surprise bills.
If AI is central to your idea, plan the limits alongside the feature. If it isn’t, it’s a natural thing to leave for later.
Offline use
Working without a connection sounds like a setting, but it means storing content on the device, recording answers while offline, and reconciling them when the learner reconnects, including when two devices disagree. Decide whether learners really study without a signal, and if so, which parts need to work offline. Often reading lessons offline is enough, which is a much smaller job than full offline review.
Admin tools
Someone has to publish lessons, fix typos and see what’s happening. That can be a full editing interface, or it can be editing the source files and re-importing. The second is far less to build, and it’s how our course sites work. Admins can also use some features that are switched off for learners, so the course team can author cards with tools the public doesn’t see.
Reporting is part of this. Our learner analytics were built entirely from data the apps already stored, with no new tracking. Planning what you’ll want to measure early keeps that kind of feature small.
How to start smaller
Most education apps get cheaper the same ways:
- One platform first, usually the web.
- One course first, in its final structure, before the whole catalogue.
- Features off, not built. Our platform turns whole features off per site. A focused course site doesn’t need user-made decks or challenges, and every feature you don’t need is one you don’t have to test or maintain.
- Gates before billing. Mark what will be paid, and add payments once there’s demand.
- Measured progress, not projections. Counts of what happened are simpler to build and easier to trust than predictions.
None of these rule anything out later. They just put the expensive parts after you’ve learned what learners actually use.
The quickest way to a real number is to tell us what you have in mind. Send us a few lines on the contact page about your content, your learners and the platforms you need, and we’ll come back with an estimate for your scope.
