"We need an app." The sentence is almost always shorthand for something else: we want to be on the home screen. We want to send push notifications. We want it to feel like a product, not a website.
None of those three goals requires an app store. That is the core of the decision — and the reason we start most projects with a Progressive Web App and only go native later, if at all.
What a PWA can do in 2026
A Progressive Web App is a website that can be installed and brings the capabilities that used to be reserved for native apps. The state of play on iOS and Android:
- Home-screen installation with its own icon, without a browser frame — on both platforms
- Push notifications — on Android for years, on iOS since version 16.4 for installed PWAs
- Offline use via a service worker caching content and requests
- Camera, location, file access, share sheet via web APIs
- One build for everything — the same codebase runs in the browser, installed, and on desktop
What a PWA cannot do: reach deep into the operating system. Bluetooth peripherals, long-running background processing, app-store billing, widgets, a listing in the store itself. On iOS some of that is restricted or only available in Safari.
What React Native brings — and what it costs
React Native produces genuine native apps for iOS and Android from one shared codebase. That gives full OS access, the app is in the store, and users experience it like any other app.
The price is operations. Two store processes with review cycles. Every update needs approval, and users do not all install it immediately — there are always several versions in the wild. Native modules have to be checked at every OS update. Reusing web components is only partly possible.
Realistically, building and running a React Native app costs two to three times a PWA with the same scope.
The decision in three questions
Does the product need something only native can do? Bluetooth, background tracking, in-app purchases via the store, AR features. If yes, the decision is made. If the answer is "not really", continue.
Is the store itself a channel? Some products get found through store search. For most B2B and service offerings that is not the case — the path runs via the website, sales or the QR code on site, and there the PWA works directly without a download detour.
Do we know yet whether the product works? If not, the most expensive mistake is building it native and then discovering users want something else. A PWA can be changed in days, not review cycles.
Why Nuxt for the PWA
Nuxt brings what a PWA needs without assembling it: server-side rendering so the product's public pages are visible to search engines, a module for service worker and manifest, routing that works the same in the browser and installed.
That is the point often overlooked: a PWA is product and marketing site at once. Landing page, login and the application itself live in one system. A native app needs a website on top — and thus two things to maintain.
We deliver the whole thing over Cloudflare, at the edge, with the timings Google measures from real user data. How we set that up technically is on the Development page.
A typical path
This is how it goes for us in most cases:
Phase 1 — PWA. Core function, login, push, home-screen install. Live in weeks. Real users, real data, real feedback.
Phase 2 — Sharpen. What gets used, what does not. Which inquiries come in. Where it sticks. The product gets sharper without a store review in between.
Phase 3 — Native, if there is a reason. If after phase 2 it is clear that a native capability or the store channel brings real value, the app is built native — with the knowledge from phase 2 instead of assumptions. Often it stays a PWA.
What that means for cost
A PWA with login, push and offline capability usually lands in the mid five figures to build, depending on scope. A comparable React Native app starts above that and brings running costs for two platforms. The concrete numbers depend on the product — what we can say about them is in What does a website cost?.
When we recommend native straight away
They exist, those cases: a product that talks to hardware. An app whose business model rests on in-app purchases. A consumer product where the store presence is part of the brand. Then React Native is the right choice, and we say so in the first conversation.
For everyone else the honest recommendation is: start with the PWA. It is faster to market, cheaper to run, and it lets you make the native decision when you know what for.
If you are facing this question right now, write to us with what the app has to do. You will get an assessment that is allowed to say "a PWA is enough".
