A recurring conversation with clients this year has been "should we build a native app?" - usually followed by sticker shock at two codebases, two app-store review processes, and two release cycles to maintain forever after. For a large share of those projects, a Progressive Web App gets most of the benefit without any of that duplication, because a PWA is not a separate technology - it's an ordinary website that has adopted three specific capabilities.

1. Served over HTTPS

Non-negotiable, and also just good practice regardless: service workers, the mechanism everything else depends on, refuse to register on an insecure origin.

2. A web app manifest

{
  "name": "Pranta Tech",
  "short_name": "PrantaTech",
  "start_url": "/",
  "display": "standalone",
  "background_color": "#fdfcfa",
  "theme_color": "#12192b",
  "icons": [{ "src": "/icons/icon-512.png", "sizes": "512x512", "type": "image/png" }]
}

Linked from the <head> with <link rel="manifest" href="/manifest.json">, this small JSON file is what lets a browser offer "Add to Home Screen," and what controls the icon, splash screen, and colours once it's launched from there.

A phone showing an installed app icon with an offline indicator
Installed from the browser, works offline, no app store in between.

3. A service worker

This is the part that actually delivers offline support. It's a script that runs in the background, separate from the page, and can intercept network requests:

self.addEventListener('fetch', event => {
  event.respondWith(
    caches.match(event.request).then(cached => cached || fetch(event.request))
  );
});

Combined with a cache populated on install, this pattern - cache-first, network as fallback - is enough to let a site's shell load instantly and keep working on a flaky connection, which matters enormously for users on inconsistent mobile networks.

Offline isn't a native-app feature anymore. It's a service worker and a cache strategy.

What you get, realistically

An installable icon on the home screen, a splash screen instead of a blank white flash on launch, offline access to previously visited pages, and - on supporting platforms - push notifications. What you don't get is deep OS integration or guaranteed app-store discoverability, so a PWA isn't a universal replacement for native. For a content site, a dashboard, or an e-commerce storefront, though, it's very often the more honest use of a client's budget.

Three things that make a website "installable"

A web app manifest (name, icons, theme color), a service worker providing at least some offline capability, and HTTPS are the baseline requirements browsers check before offering an install prompt. None of them require rewriting the app - they're additive to an existing site.

{
  "name": "Pranta Tech Journal",
  "short_name": "PT Journal",
  "start_url": "/",
  "display": "standalone",
  "icons": [{ "src": "/icon-512.png", "sizes": "512x512", "type": "image/png" }]
}

Where PWAs genuinely replace a native app, and where they don't

Push notifications, offline access, and a home-screen icon cover a large share of what most businesses actually need a "mobile app" for - a news reader, an ordering app, an internal tool - at a fraction of the cost of two native codebases. Deep OS integration (Bluetooth, background location, complex camera pipelines) still genuinely needs native code; knowing which category a project falls into up front saves a lot of wasted native development.