What an app costs in 2026
The question has no single answer, it has three, because there are three different ways to buy the same thing. And the gap between them is not ten or twenty per cent: it is two orders of magnitude.
This guide separates the cost of building, of publishing and of maintaining, which is the part that usually goes missing.
The three routes, and what each one charges
An agency. You buy a project: discovery, design, development, testing and delivery. The price is team hours times scope, and quotes for a business app run from tens to hundreds of thousands. It is the most expensive route and the only one where somebody is contractually answerable for the result.
A freelancer. The same work with one person. Well below an agency, with a different risk: continuity. The code is yours, but the reasoning behind each decision leaves with them.
A platform. You are not buying hours, you subscribe to a tool and build it yourself. The price is fixed and monthly, does not grow with scope, and is known before you start. On Fabapp: Free to build and publish on the web, and paid plans from R$ 99 to R$ 1,099 a month.
One number is not comparable to the other. Agency and freelancer are a delivery price; a platform is a tool price, and your own time enters as an unbilled cost.
What pushes a quote up
Not every feature costs the same. Four things dominate, and three of them never show on screen:
- Sign-in and user accounts. The moment the app stops being a site and starts having people inside it. It brings password recovery, permissions and the question of who sees what.
- Payments. Card, recurring subscriptions, refunds, a charge that fails. It is not a button, it is a flow with states.
- Store publishing. A developer account, binary signing, a privacy policy, a data form and review. That is process work, not code.
- An admin area. Almost every quote forgets it and almost every app needs it: somebody has to see the orders, edit the products and check the takings.
The cost nobody budgets: keeping it running
A published app costs money even when nothing changes:
- hosting and a database, which grow with usage;
- a domain and a certificate, yearly;
- the Apple developer account, US$ 99 a year, charged for the app to stay in the App Store;
- technical upkeep, because libraries age and stores move the goalposts (the minimum Android target version changes every year);
- support, which is somebody's hour when a customer cannot sign in.
Budget only the build and you meet this in year two, which is where a lot of published apps die.
How to decide without guessing
Three questions settle most cases:
Is the app the product, or a tool of the business? If it is the product, and the edge is something technical nobody else does, you will need people. If it is how you sell, book or serve better, a platform handles it.
Do you need the stores, or is the web enough? An app installable from the browser skips review, has no yearly fee and ships the same day. Read native app, PWA or website before assuming you need a store.
What is the passing month worth to you? The most expensive route is usually also the slowest. If the difference between launching in a week and launching in four months changes your business, that is part of the price.
The short version
- building: an agency is dearest, a freelancer sits between, a platform is the only one with a fixed, predictable price;
- publishing: US$ 25 once with Google, US$ 99 a year with Apple, and none of it depends on where the app was built;
- maintaining: always exists, and is the cost that almost never appears in a quote;
- the genuinely cheapest app is the one you can change yourself once it is live.
For the numbers on your own case, the per-segment guides do the maths: delivery, transport and a simple app.