Why this article now

PWAs have had a roller-coaster trajectory. Launched by Google in 2015, long held back on iOS, they became in 2023-2024 a credible option for nearly all consumer mobile projects. In 2026, PWA capabilities on iOS and Android are sufficient to replace a native app in 90% of cases — for 3 to 10 times less.

This article tours the actual capabilities, residual limits and use cases where the PWA is the right answer.

What a PWA is, concretely

A PWA is three technical ingredients added to a classic website:

  1. A manifest (JSON file) that describes the application: name, icon, splash screen color, installation behavior
  2. A service worker: a script that runs in the background, manages cache, offline, notifications
  3. Mandatory HTTPS: the PWA only installs on a site served over HTTPS

When a visitor lands on your PWA site from their phone, the browser offers: "Add to home screen". A tap, and the app installs: icon on screen, full-screen launch, native-app-like behavior.

PWA capabilities in 2026

Install and app experience

  • Icon on home screen ✓
  • Full-screen launch (no address bar) ✓
  • Splash screen at startup ✓
  • Standalone behavior indistinguishable from a native app for the end user ✓

Connectivity

  • Full offline mode via service worker (asset caching, data, reconnection management) ✓
  • Background sync when connection returns ✓
  • Offline-first cache for apps that must work in degraded situations ✓

Sensors

  • Geolocation (position, configurable accuracy) ✓
  • Accelerometer, gyroscope, orientation ✓
  • Camera (photo capture, QR code scan, file selection) ✓
  • Microphone (audio recording) ✓
  • Biometric sensors: limited (Touch/Face ID via WebAuthn) — possible with constraints
  • Advanced sensors (heart rate, NFC, deep Bluetooth): limited or unavailable

Notifications

  • Push notifications on Android: fully supported for a long time ✓
  • Push notifications on iOS: supported since iOS 16.4 (March 2023). Small constraint: the user must install the PWA on the home screen to be entitled ✓
  • Local notifications (triggered by PWA code) ✓

Performance

  • As fast as a native app on 95% of uses
  • Smooth animations at 60 fps possible with careful design
  • Limit: 3D games, real-time video editing or augmented reality remain better served by native

Data persistence

  • LocalStorage: for small simple data
  • IndexedDB: for larger bases, high-performance queries
  • Cache API: for network resources
  • Storage capacity: several gigabytes depending on the browser

Uses where PWA shines

Field tool

A field intervention team needs a mobile tool: report entry, photos, geolocation, working even in no-network areas. → Perfect PWA. No-store install (simpler centralized management), offline operation, sync on reconnection.

Treasure hunt, tourist trail, event

The user arrives on-site, scans a QR code, installs the app in one tap. Geolocation, contextual content, local persistence to track progress. → Ideal PWA (see Hermitage Jeu de piste).

Booking app, appointment scheduling

Classic website with client area + installable version for regular users (reminder notifications, home screen shortcut). → Perfect PWA, no significant extra cost vs the website.

Mobile-first marketplace

A mostly-mobile marketplace site can be offered as an installable PWA. Frequent users install it, others use it as a website. No native needed as long as there's no mandatory in-app payment.

Internal tool for collaborators

For 50, 500 or 5,000 internal collaborators, deploying a native app demands managing deployments via Apple Business Manager / Google Workspace. A PWA installs via a simple link — radically simplified management.

Residual limits to know

1. Not in the stores

For a consumer product that depends on store visibility, it's a real limit. For B2B, internal tools, usage captured by your direct marketing: it's irrelevant.

2. No appearance in some iOS "natives"

A PWA installed on iOS appears on the home screen, but not in the native app switcher (Home double-tap to switch between apps). It's a UX detail that goes unnoticed for most users.

3. User confirmation for install

Installing a PWA requires an explicit user action (add to home screen). It's less smooth than clicking "Install" in a store. Mitigation: onboarding that guides this step, especially on iOS where the gesture is less obvious.

4. High graphic performance

For uses pushing display to its limits (3D games, complex animations, real-time video), a PWA remains slightly behind native. For 95% of business uses, no perceived difference.

5. Deep OS integrations

Home screen widgets, Siri / Google Assistant integration, iOS App Clips, advanced sharing via Action Extensions: these are native features a PWA doesn't have (or doesn't yet).

User-side installation — UX to refine

It's the main UX challenge of a PWA. On Android, the browser often proposes installation spontaneously; on iOS, you must guide the user ("Tap Share → Add to home screen").

Patterns that work

  • Small visual tutorial on first mobile visit: 3 images showing the 3 gestures
  • Non-intrusive banner in the footer: "Install this app on my phone"
  • Platform detection: show only on mobile, adapt message to iOS / Android

Patterns that fail

  • Aggressive modal on first visit ("INSTALL NOW!") → user closes and leaves
  • Multiple reminders: insisting kills conversion
  • Technical text: "PWA", "manifest", "service worker" mean nothing to the end user

The "half-PWA" trap

Many sites claim PWA status because they have just a manifest. For a PWA to hold its promise, you need a well-designed service worker that:

  • Caches essential assets at first load
  • Manages offline mode with a clear strategy (network-first, cache-first, stale-while-revalidate by resource)
  • Synchronizes queued modifications when network returns
  • Updates cleanly when a new version is deployed

Without that, the "PWA" delivers nothing more than a responsive mobile site.

Maturity check — is your project PWA-friendly?

Give yourself a score out of 5 (one point per "yes"):

  1. Is the main usage on mobile?
  2. Do your users come back regularly (at least a few times per month)?
  3. Is geolocation or offline mode useful to your use?
  4. Are you ready to recruit your users through your direct marketing (not via the stores)?
  5. Can you do without mandatory in-app payment through the stores?
  • 4-5 points: PWA is very likely the right answer. Major savings vs native.
  • 2-3 points: PWA possible as MVP, to reassess after first user feedback.
  • 0-1 point: reconsider the real need for a mobile app. A responsive mobile site may be enough.

Going further

You have a mobile project to scope? Run the diagnostic — I'll orient you honestly between PWA and native.