Ginza · Lovable to the App Store in 3 weeks

The starting point

Ginza had product-market fit as a web app · a growing Brazilian community, hundreds of active users, daily posts. What it didn't have was a place on its members' home screens. Mobile web sessions maxed out at 2–3 minutes. Push was impossible on iOS Safari. The share of returning weekly users was flat. Every retention lever pointed at the same missing piece: a real native app.

Why not a rewrite

A React Native or Swift rebuild would have thrown away every line of Lovable code already shipping value · 3–6 months, tens of thousands of pounds, and a parallel codebase to maintain forever. A DIY WebView wrapper (Natively, GoNative) was the other extreme · cheap, but statistically likely to be rejected under Apple Guideline 4.2, and missing the exact features that make a social network sticky (real push, biometrics, deep links).

The right answer sat in the middle · a done-for-you Capacitor shell with every native plumbing point actually wired up. That's the KW Native process.

What we shipped

1. Firebase Cloud Messaging push

   Real APNs push · not web push. Every new message, comment and follow arrives on the lock screen. This is the single biggest driver of Ginza's retention curve · users with push stay past 11 sessions, users without are gone after 2.

2. Google + Apple OAuth

   Apple Sign In is mandatory under Guideline 4.8 for any app that offers third-party login · a huge share of WebView rejections come from missing this. Ginza had both wired up before first submission.

3. Face ID / Touch ID

   Biometric quick-login on every launch · the difference between a social network you open and one you don't.

4. Universal Links

   ginza.com.br links open the app when installed, the mobile site when not · no browser hand-off, no lost session, no broken share flow.

5. Microphone permissions

   Voice notes work natively · a permission surface that trips up WebView wrappers on first review.

6. Live-URL auto-sync

   The native shell loads the live Lovable-hosted web app · every deploy to ginza.com.br is an instant app update. No rebuild, no resubmission for content changes.

The migration · 355 users, zero data loss

The existing Supabase database moved across intact · every account, post, follow-graph and media asset preserved. Members opened the new app already logged in, with the same feed they left the mobile site on. The switchover ran in a single evening; no member noticed the plumbing.

First-review pass · Guideline 4.2 cleared

Apple rejected 25% of all App Store submissions in 2024 · Guideline 4.2 (Minimum Functionality) is the single largest cause, and it's where every WebView wrapper trips. Ginza cleared review on the first submission because the native plumbing genuinely exists · real APNs push, Apple Sign In, Face ID, deep links, microphone permissions. Every feature Apple checks · we already had.

The timeline

Why Ginza matters to your project

Every WebView-wrapper tool on the market has customers stuck in rejection loops · read Natively's own Trustpilot and Apple's developer forums. Ginza is the counter-proof. Same Capacitor runtime, same submission process, same team · already live and passing Apple review with the exact features (real push, OAuth, Face ID, deep links) that cause the rejections everyone else hits. When KW Native quotes you a 10-business-day delivery, Ginza is the receipt behind that number.