Back to Case Studies

Building and Maintaining a Cross-Platform Mobile App

Grew a mobile product to more than 50,000 lifetime sales on iOS and Android, rebuilding it as platform needs changed and keeping it current for years.

Mobile AppsFlutterDartApp StoresProduct Analytics

The business problem

A mobile product only pays off if it keeps working. Operating systems change, store requirements tighten, and a codebase built for one platform does not automatically serve another. Turning an idea into a product people keep buying means choosing what to build, shipping it on the right platforms, and maintaining it long after launch.

Focusing on what worked

The business started as a set of about 20 apps. Sales data made the decision easy: 19 of them were unprofitable, so they were unpublished and development concentrated on the one product customers actually paid for. Less surface area meant more attention on quality, support, and updates for the product that mattered.

Rebuilding for new platforms

The product was rebuilt twice as the platforms changed. The first rebuild reworked the Android version end to end and added iOS, so the product reached both app stores. A later migration to Flutter and Dart put both platforms on a single codebase, which cut the effort of keeping them in step.

Content lives apart from code. Companion Python tooling handles collecting, deduplicating, and ingesting content, so routine content updates do not require reworking the app itself.

Letting evidence set the direction

Decisions about the product come from evidence rather than instinct. Anonymous usage analytics in PostHog, with nothing collected that identifies a customer, show which features people use, how quickly each release is adopted, and whether changes help customers reach the goal they bought the app for.

Store reviews are pulled from the App Store and Google Play through their APIs at the start of every release. They surfaced something the usage data could not: after a major update grew the product’s content library, customers found it overwhelming and asked for the focused original back.

The answer was analysis, not deletion. A Python analysis recovered the original content set from version history, grouped near-duplicates, checked each item against published reference material, and ranked everything by how well it serves the customer’s goal. That lets the product lead with a focused core set while keeping the rest available, and an upgrade test built from an earlier version’s saved progress confirms that no customer loses their work in the change.

Operating it like a business

Running the product includes more than writing code. Store submissions and reviews, Apple and Google developer accounts, payment processing, customer support, and ongoing compatibility updates are all part of delivery. A typical year needs only occasional compatibility and content updates, because the scope and structure were designed to be maintainable.

The result

The app has passed 50,000 lifetime sales across the App Store and Google Play. That figure counts purchases, not downloads or revenue. It continues to sell largely through organic store search, with a small, predictable maintenance workload.

The same approach applies to client work: ship something focused on a real need, measure how people actually use it, choose a structure that is cheap to maintain, and keep it current as the platforms around it change.

Explore more case studies

See how software, automation, and infrastructure address different business needs.

View Case Studies