Applications

PWA or Native App: The Comparison I Use to Decide

Installation, updates, offline use, notifications, device access, maintenance cost: a comparison table and the cases where a PWA is enough and where it isn't.

When I talk with someone about an app, the conversation almost always narrows to two options: an app in the store, or "just a website". In between there's a third option that many people don't know or don't take seriously: the progressive web app, PWA. I want to compare it with a native app here without cheering for either, because I've chosen both and had a reason each time.

In Does Every Business Need an App? I wrote about when an app makes sense at all. Here I assume it does, and deal with the question of which kind.

Short answer

A PWA is a website the user adds to their home screen that then behaves like an app: it opens without the browser frame, works offline for whatever has been cached, and can send notifications. A native app is a program written for iOS or Android and installed from the store. For most business cases, where the user needs quick access, forms and notifications, a PWA covers the need at lower cost. A native app is justified when you need deep access to the device, the best performance, or a store presence as part of your strategy.

Side by side

PWA Native app
Installation From the browser to the home screen, no store From the App Store / Google Play
Updates Immediate, one version for everyone Through the store; users on different versions
Offline Yes, for pre-cached content Yes, unrestricted
Push notifications Android yes; iOS from 16.4, only when added to the home screen Yes, unrestricted
Device access Camera, location, sharing, basic sensors; limited for Bluetooth, NFC, background work Full
Performance Good for forms, lists, content; weaker for 3D, video processing Best
One codebase for all platforms Yes Usually two, or a cross-platform framework
Discoverability Like a website, in Google In the store and via links
Maintenance cost Like a website Higher: OS versions, store rules, reviews
Pre-release review None The store reviews every version

Where a PWA is enough

For business apps that people use to enter data, look things up and receive notifications, a PWA in my experience covers almost everything. Bookings, ordering, a customer portal, time tracking, a catalogue. The advantage clients underestimate most is updating: I fix a bug once and it's fixed for everyone, with no waiting for review and no users who never install the update.

The second advantage is that a PWA isn't a separate project. It's the same code as the website, plus a manifest, a service worker for caching, and some thought about what should work offline. For the client that means one thing to maintain instead of two or three.

Where a PWA isn't enough

A few cases where I'd choose a native app without hesitation:

  • Hardware access the browser doesn't give you. Bluetooth devices, NFC, background work, precise location while the app is closed.
  • Heavy graphics or processing. Games, video editing, 3D.
  • Notifications on iOS as the core feature. They work in a PWA too, but only after the user adds the app to the home screen, which most people don't do on their own. If the notification is the heart of the product, that's too big a hurdle.
  • The store as a channel. If you expect people to look for you in the App Store, or if being in the store is part of trust in your industry.
  • In-app payments under store rules. That's a separate topic with its own rules and commissions.

Support I check on a real device

PWA support differs between browsers and operating systems, and it changes. Before deciding, I pick up a phone and check: does the app install, does the notification arrive, does it work offline, does the camera work. A few minutes with a real device tell you more than an hour reading support tables. For iOS, the caveat in the table above applies: according to WebKit's documentation, web push is available from iOS 16.4 only for apps added to the home screen.

The middle path

Often the decision isn't "either–or" at all. I start with a PWA because it's cheaper and faster, and if it turns out that users really do use the app every day and are missing something the browser can't provide, the native app gets built on evidence rather than guesswork. The reverse path is more expensive: a native app nobody installs is a cost with no feedback.

How I decide

The questions I ask are simple: what will the user do, how often, on which device, and what of that can the browser not do. If the answer to the last question is "nothing", it's a PWA. If the answer is concrete, it's a native app, and I know what I'm paying for.

Photo of the author

I build web solutions and digital products. At GE-KO I work on web design, development and SEO; with Bizko I look for simpler ways through everyday business. I am interested in how technology saves people work, and in how AI is changing the web.

One step further.