No App in the Store, Still Offline: The PWA Question
For internal line-of-business apps in SMBs, the PWA is almost always the smarter choice. In 2026 the last 'but you can't do that' arguments are gone.
"We need an app" means, in 80 % of cases, "we need something that works offline, installs on a phone, and can send push notifications." The Progressive Web App has offered exactly that for years - and in 2026 the last excuses are gone.
What Changed in 2025/2026
Three gaps that PWA articles still listed in 2022 have closed:
- Push on iOS: officially supported since iOS 16.4, stable on screen-sleep since iOS 17.
- Install UX: Android offers active WebAPK invitations, iOS exposes visible share-menu actions, Chrome/Edge one-click install on desktop.
- File system access: File System Access API covers Chrome/Edge-based scenarios; for iOS, download/upload remains sufficient for most business flows.
When the PWA Is the More Honest Choice
These questions weigh in favour of PWA:
- Should field or service staff work offline but sync back at the office?
- Does the app run on company hardware (iPad, Android tablet, laptop)?
- Is the app store overhead (Apple developer account, review, sign-off) a blocker?
- Should updates reach every device instantly, without review delays?
If three of four apply, the PWA is the more direct path. Native remains the right call for B2C distribution, deep hardware integration (Bluetooth profiles, NFC host-card emulation), or apps that need strict store features (StoreKit, Apple Watch).
What Works Pragmatically on Real Projects
A reliable setup from our projects:
- A lean JavaScript framework plus a service worker for offline cache.
- IndexedDB for structured offline data.
- Background sync for deferred uploads once connectivity returns.
- Web push for notifications, with clear opt-in moments.
The decisive factor is not the framework but offline discipline: which data must be offline, which writes are queued, how conflicts are resolved.
A Concrete Example: QSOlog
What this looks like in practice: our open-source project QSOlog - an offline-first logbook for amateur radio operators. Vue 3 + Vite on the frontend, Dexie on IndexedDB for structured offline data, installable on desktop and mobile, no app store involved. After the first load everything runs locally: QSO entry with automatic band detection, map visualisation, ADIF/CSV/JSON export, PDF reports. A textbook case where the PWA is the more honest answer to "we need an app."
Bottom Line
In 2026 the PWA is not a compromise - for internal line-of-business apps it is often the technically and organisationally better answer. "We need an app" should open a conversation, not close it.