When a company decides to build a mobile app, the first technical debate is almost always the same: build separately for iOS and Android, go with Flutter or React Native, or skip all that and ship a PWA. Each option has its champions, and each is right in its own situation. The problem is that the choice is often made on the basis of "that's what our contractor knows" rather than what the product needs.
Below is an honest comparison of approaches to mobile app development for business: the strengths and weaknesses of native, cross-platform and PWA, when to choose which, and what you need to know about publishing in the App Store and Google Play. No holy wars, just the criteria we use to make the call on our own projects.
The problem first, then the technology
Before choosing a stack, answer a few questions. The answers matter more than any comparison table.
- Why an app if you already have a website? Repeat purchases, a loyalty program, push notifications, offline use, access to the camera, location or Bluetooth. If you need none of that, you may not need an app at all.
- Who are the users, and on which devices? Customers of a mass-market service, warehouse staff with company Android terminals, drivers, partners.
- How much does a native feel matter? For a banking or media product, where users compare you with the market leaders, it's critical. For an internal inventory tool, it's secondary.
- What about the device? Advanced camera work, AR, background location, wearables integration, low-level networking.
- Who will maintain the app two years from now? An in-house team or a contractor, and what skills they have.
We work through these questions in the first stage, audience and user flow research. We choose the technology after that, not before.
Choose the stack based on the product's needs, not on "that's what our contractor knows."
Native development: the full power of the platform
A native app is written separately for each platform, in Swift for iOS and Kotlin for Android, using each platform's own tools and UI components.
- Pros. Full, fast access to every device capability and to new OS features as soon as they ship. The best performance in demanding scenarios: complex animation, video processing, camera and sensor work. The interface behaves exactly the way the platform's users expect.
- Cons. Two codebases mean, in effect, two teams or two workstreams, and every feature gets built twice. Development and maintenance cost more, and it's harder to keep releases in sync so the iOS and Android versions don't drift apart in functionality.
When to choose it. A product where the app is the main channel and a key competitive advantage; heavy hardware work; high demands on performance and smoothness; ample resources for long-term development.
Cross-platform development: Flutter and React Native
Cross-platform development lets you write a single codebase for iOS and Android. The two most widely used frameworks work differently.
Flutter
Google's framework, written in Dart. It renders the interface with its own engine, so the app looks the same on every platform and you control every pixel. It suits products with an expressive custom design where you don't need to replicate the iOS and Android system components exactly. The limitation: there are fewer developers on the market than for JavaScript, and device-specific features sometimes require native modules.
React Native
Meta's framework, written in JavaScript or TypeScript. It uses the platforms' native components, so the interface feels closer to the system. A big plus is the shared language and some shared logic with web development: a team that builds your website or web service in React can reuse its knowledge and part of the code. The limitations are similar: unusual device work needs native modules, and framework and library updates need attention.
Where cross-platform falls short
Heavy on-device computation, complex graphics, deep integration with system features that have only just shipped. Another risk is dependence on third-party libraries: if the library for a feature you need is abandoned, you'll have to take over its maintenance.
When to choose it. Most business apps: online stores, booking and delivery services, customer accounts, loyalty programs, internal tools. One team, one codebase, synchronized releases and much lower maintenance costs, with a high-quality interface.
PWA: an app without the app stores
A PWA (Progressive Web App) is a web app that can be installed on the home screen straight from the browser. It can work offline with cached data and send push notifications.
- Pros. One codebase for every device, desktops included. No store publishing or review: updates reach users immediately. It opens from a link and is indexed by search engines. If you already have a web service, a PWA can often be built from it with relatively modest changes.
- Cons. Limited access to device features, especially on iOS: some APIs aren't available, and push notifications there only work for a PWA added to the home screen. Users can't find the app in a store and don't always understand how to "install" it. For a mass-market consumer product, that's a real loss.
When to choose it. Internal tools for employees, partner services, quick hypothesis testing before investing in a full app, products where most traffic comes from search and links. The project logic here is closer to building a web service than to mobile development.
How to choose: a quick cheat sheet
Put the three approaches side by side, and the differences look like this.
| Native | Cross-platform | PWA | |
|---|---|---|---|
| Codebase | Two: iOS and Android | One for iOS and Android | One for all devices, desktops included |
| Device access | Full, new OS features right away | Broad; native modules for specific features | Limited, especially on iOS |
| Development and maintenance | More expensive: every feature is built twice | Much cheaper maintenance, synchronized releases | Can often be built from an existing web service |
| Publishing | Through the stores, with review | Through the stores, with review | No stores, instant updates |
| When to choose | The app is the main channel; heavy hardware work | Most business apps | Internal tools, partners, hypothesis testing |
- Need to test an idea quickly and cheaply: a PWA or a cross-platform MVP.
- A typical customer-facing business app: cross-platform, Flutter or React Native depending on the design and the skills of the team that will maintain it.
- You have a React web team and logic shared with the website: React Native.
- A rich custom visual style and animations: Flutter.
- Advanced camera, sensor, AR or background work: native.
- An internal tool for employees: a PWA or cross-platform, depending on whether you need device features.
For example, if the brief is an app for hotel or resort guests, what matters is working on the slopes with a weak signal, location, and buying services in a couple of taps. For a mobile VPN service like the one we built, the platform's system-level networking capabilities and reliable background operation come first. Different problems, different priorities when choosing a stack.
What to build in from the first release
Whichever approach you choose, a mobile app is more than screens on a phone. Some things are cheaper to plan for up front than to add later.
- Backend and admin panel. The catalog, promotions, copy and push campaigns should be managed without shipping a new version. Otherwise every content change waits on store review.
- Analytics and crash reporting. Without them, you won't know at which step users abandon checkout or on which devices the app crashes.
- Forced updates. A mechanism that asks users to update when an old version no longer works with the server.
- Deep links. So that a link in an email, ad or text message opens the right screen in the app, not the home screen.
- Account deletion and personal data handling. This is a requirement of the stores and of data protection law such as GDPR, not a nice-to-have.
The scope of the first release also shapes the budget. The more precisely the user flows and integrations are described, the more accurate the estimate.
Publishing in the App Store and Google Play
Publishing is a separate stage that's often underestimated in project timelines.
- App Store. You need an Apple Developer Program account, ideally registered to the company. Every version goes through manual review against Apple's guidelines. Common reasons for rejection: unfinished functionality, no way to delete an account, unclear payment flows for digital goods, insufficient disclosure of data collection.
- Google Play. Review is usually more lenient, but there are requirements of its own: the data safety declaration, a privacy policy, an up-to-date target Android API level. New personal developer accounts must run a closed test before releasing to production.
The work continues after publishing: updates for new OS versions, bug fixes, new functionality driven by analytics. The principle is the same as for a website: decide in advance which user actions count as success, and track them as conversions and events.
FAQ
Which is cheaper: native or cross-platform development?
Usually cross-platform: one codebase instead of two, one team, synchronized releases. The difference is most noticeable in maintenance and ongoing development, where every native feature has to be built twice.
Flutter or React Native: which should we choose?
Flutter suits products with a rich custom design and animations, since it renders the interface itself. React Native pays off if the company already does web development in React and wants to reuse skills and some of the logic. Both work for most business apps.
Can a PWA replace a mobile app?
For internal tools, partner services and testing hypotheses, often yes. For a mass-market consumer product, a PWA falls short: it isn't in the stores, device features are limited on iOS, and push notifications only work after it's added to the home screen.
Do we need to publish in both the App Store and Google Play?
If your audience uses both iOS and Android, yes. Each store has its own developer account, review process and requirements, so plan the time for preparing builds and listings for both. Cross-platform development makes it easier to keep the two versions in sync.
Whose name should the store developer accounts be in?
The client company's. That way the app, its reviews and its publishing history belong to you rather than the contractor, and changing teams won't cost you access to the app.




