KMP answers a different question from the frameworks of Cordova-Capacitor. They ask how one team can ship to every store from a web skill set; KMP asks how a team that already writes native Kotlin can stop writing the same logic twice. Measured against this book's other routes:
| Capacitor 243,123 (4) | React Native 36,878 (3) | KMP, shared logic | KMP + Compose 234 UI | |
|---|---|---|---|---|
| Language | JS/TS + web UI | JS/TS + React 7,897 | Kotlin + Swift | Kotlin |
| UI on iOS | WebView | Native views | SwiftUI, by hand | Compose on Skia 351,502 |
| Code shared | Nearly all | Most | Data and rules | Nearly all |
| Runtime cost | WebView | Hermes 11,328 | None | Skia engine |
It is worth it when you already have a Kotlin Android app and an iOS app that must behave identically: pricing rules, sync, validation, and API models are where two hand-written versions drift apart, and where one tested module pays off first. Google officially supports KMP for sharing business logic between Android and iOS, and Jetpack 234 's Room 234 , DataStore 234 , ViewModel, Navigation and Paging, as well as Ktor 274,728 , Koin and Coil 11,909 , all run in common code.
It is weaker for a team of mostly iOS developers, who must read Kotlin, debug across the language boundary, and accept an API shaped by the Objective-C header while Swift export is Alpha. Android-only libraries such as Hilt, Glide 35,023 and CameraX 234 stay in androidMain behind interfaces. And Compose Multiplatform 25,990 for iOS, stable since 1.8.0 (May 2025), does not look exactly like UIKit.
For BookNest, the sensible path is the one this section built: share the model and catalog rules, keep Jetpack Compose on Android, and let an iOS team choose SwiftUI or Compose Multiplatform later.