All posts

PWA or native? Your users decide.

The PWA vs native app choice comes down to one question: does your product need deep access to the phone? If not, a progressive web app gets one version to every device, without app store review. If it does, native is worth the extra cost. Here is how to tell, and how to avoid building twice.

For founders choosing
how to ship their app.

One build, or the whole device.

PWA vs native app: the short answer for most first versions

A progressive web app, or PWA, is a website built to behave like an app. People open it from a link, add it to their home screen, and it starts full screen with its own icon. It can work offline, send notifications and store data on the device. One codebase runs on iPhone, Android and desktop.

A native app is built specifically for iOS or Android with the tools each platform provides, and it is installed from an app store. It gets full access to the camera, sensors, Bluetooth and background tasks. You usually pay for that with more build time, store review and, unless you use a cross-platform framework, two codebases.

For most first versions, a PWA is the sensible default. It reaches users fastest, updates instantly and costs less to change while you are still learning what people want. Native becomes the better choice when a core feature depends on the device itself, or when being found and trusted in an app store is part of how you win customers.

  • Choose a PWA when the product is mostly screens, forms, data and accounts.
  • Choose a PWA when people arrive through links, search or a QR code.
  • Choose native when a core feature needs sensors, Bluetooth, NFC or long background work.
  • Choose native when app store presence or in-app subscriptions are central to the business.

Pick wrong, pay twice.

Why the wrong platform choice leads to a second build

The platform decision is usually made early. Founders hear that real apps live in the app store, so they commission two native apps for a product that is really a booking form with a dashboard. Months later the idea needs to change, and every change has to be made, tested and reviewed twice.

The opposite mistake is just as expensive. A team builds a web app for a product that depends on scanning, precise location in the background or a connected device, then finds out on an iPhone that the browser simply does not allow it. The workaround is a rebuild, at the moment the product should be gaining users.

Both mistakes have the same cause: the choice was made on assumptions about technology instead of on what users must be able to do. Once you start from that list, the answer is usually obvious.

Start with what users must do.

What to decide before choosing between a PWA and a native app

Write down the three to five things a user must be able to do in your first version. Then, for each one, note which part of the phone it touches. Logging in, browsing, filling in forms, paying for physical goods and receiving notifications all work well on the web. Continuous background location, Bluetooth hardware, NFC, LiDAR scanning and heavy on-device processing generally do not, at least not reliably on every platform.

Next, look at how people will find and use the product. If they arrive from a newsletter, a search result, a poster or a colleague’s link, a PWA removes the step of visiting a store and installing something. If your audience expects to search an app store, or the app will be used daily for years, a store listing earns its place.

Finally, check the business rules. App stores have their own rules for selling digital content and subscriptions inside an app, and every release goes through review. A PWA avoids those rules but also misses the store’s payment and discovery features.

  • The three to five tasks your first version must support.
  • The device features each task needs, if any.
  • How users will discover the product and how often they will open it.
  • Whether you sell physical goods, services or digital content.
  • Requirements for offline use, privacy and where data is stored.

Same backend. Different front door.

How a progressive web app and a native app differ technically

A PWA is a web application with two extra pieces. A manifest file tells the phone its name, icon and how to open it. A service worker runs alongside it, caches files so the app opens without a connection and receives push messages. Data lives on your server and, where useful, in browser storage on the device. It must be served over HTTPS.

The limits sit in the browser, and they differ per platform. Android browsers support most modern web features, including install prompts. On iPhone, every browser uses Apple’s web engine, people must add the app to the home screen themselves, notifications only work after they do, and background tasks and hardware features such as Bluetooth and NFC are not available to web apps. Check the current state of each feature before you depend on it, because it changes.

A native app talks to the operating system directly. That is what makes sensors, background processing and platform features dependable. It also means separate builds, store review for every release, and users on older versions for months.

What does not change is the backend. Accounts, permissions, business rules, payments and integrations live on the server behind an API, whichever front end you choose. AI features usually run there as well, calling a hosted model, so they behave the same in a PWA and a native app. Build that layer well and a later move to native means replacing the front end, not starting over.

Ship one. Then earn the next.

A safe path from first version to the right platform

Start with a short scoping round: the user tasks, the device features they need and how people will reach the product. If nothing on that list requires native access, build the first version as a PWA with a clean API behind it. If one core task does, test that task on real phones early, before the rest is built around it.

Test on the devices your users actually carry, not only on a laptop. Install the app on an iPhone and an Android phone, turn off the network, deny a permission and see what happens. Check sign-in, data storage and access rights before real users arrive, especially when personal data is involved.

After launch, let usage decide the next step. If people return daily, ask for features the browser cannot offer or expect a store listing, that is evidence for a native or cross-platform app. Because the backend already exists, that next version reuses everything except the screens.

  • Scope the user tasks and the device features they rely on.
  • Prove any risky device feature on real phones first.
  • Build the first version on a clean, documented API.
  • Test install, offline use and permissions on iPhone and Android.
  • Add a native app only when usage gives you a reason.

How we approach it.

How Walhallah chooses between a PWA and a native app

We start from what your users need to do, not from a preferred technology. We have built both kinds of product. Ommetje met Tom and Get Sett are progressive web apps that people open straight from a link, while WalkerScan is a native iPhone and iPad app, because scanning spaces with LiDAR needs direct access to the device.

For an MVP, we usually recommend the simplest platform that supports the core tasks, with AI at the core of how we build so the first version reaches users in weeks rather than months. We design the backend so it can serve a native app later without a rewrite. And if your idea genuinely needs native from day one, we will say so before you spend money on the wrong version.

Questions. Answered.

Is a PWA cheaper to build than a native app?

Usually, yes, because one codebase serves every device and there is no store review for each release. The real cost drivers are the same for both: the number of screens and user roles, the complexity of the rules and the integrations. Native adds work per platform and per store.

Can a PWA be listed in the Google Play Store or the App Store?

A PWA can be packaged for the Google Play Store with a thin wrapper, which many teams do. Apple’s App Store expects apps that offer more than a wrapped website, so a pure web app is often rejected there. If an iPhone store listing matters, plan for a native or cross-platform app.

Do push notifications work in a PWA on iPhone?

Yes, but only after the user adds the web app to their home screen and grants permission. On Android, notifications work more broadly. If notifications are central to your product, design an onboarding step that explains how to install the app, and test it on real devices.

What about cross-platform frameworks instead of fully native?

Cross-platform frameworks let one codebase produce real iOS and Android apps with access to most device features. They are a strong middle ground when you need the app stores but want one team and one codebase. Very specialized features, such as advanced camera or sensor work, can still require platform-specific code.

Can we start with a PWA and switch to a native app later?

Yes, and it is often the most sensible path. If accounts, data, rules and integrations live behind an API, a native app can reuse all of them. You rebuild the screens, not the product.

Web app or native? Let’s find out.

Tell us what you’d like to build or improve.