OnTheBeach.dev presents · 03 chapters

Building a mobile app today.

From the first need to ongoing improvement: the process, technologies and role of artificial intelligence.

Explore
The big picture

Three perspectives on one product.

  1. 01The stages of a mobile project
  2. 02Technology choices
  3. 03AI in the process
01 / The journey

An app takes shape long before the first line of code.

Nine stages from the initial need to ongoing product development. Each decision informs the next.

01
Stage 01 / 09

Understanding the need

Understand the real problem before discussing solutions: users, use cases and constraints.

  • Workshops with stakeholders and target users
  • Personas, journeys and priority use cases
  • Functional scope: MVP and long-term vision
02
Stage 02 / 09

Business model review

Check that the project has a viable business model before investing in development.

  • Business model: freemium, subscription, purchase or advertising
  • Customer acquisition cost versus lifetime value
  • Development and operating budget
03
Stage 03 / 09

Design

Turn needs into a tangible experience: journeys, screens and technical architecture.

  • Wireframes followed by detailed UX/UI designs
  • Technical architecture and backend choices
  • Specifications shared with the whole team
04
Stage 04 / 09

Planning

Break the project into deliverable releases and set a realistic pace.

  • Deliverable releases and sprints
  • Effort estimates by feature
  • Plan dependencies: design, APIs and app stores
05
Stage 05 / 09

Development

Build the app following the agreed architecture and priorities.

  • Continuous integration and code reviews
  • Follow Apple and Google guidelines
  • Front-end and back-end communication through documented APIs
06
Stage 06 / 09

Testing

Verify that the app meets user needs without breaking existing features.

  • Functional and regression testing
  • Checks on real devices and operating systems
  • User acceptance testing before release (UAT)
07
Stage 07 / 09

Launch

Publish the app in stores for its first users.

  • Submit to the App Store and Play Store
  • GDPR and store policy compliance
  • Gradual rollout to limit risks
08
Stage 08 / 09

Monitoring

See how the app performs once it is live.

  • Technical monitoring and crash reports
  • Analysis of real-world usage (analytics)
  • Listen to reviews and user feedback
09
Stage 09 / 09

Evolution

Keep improving the product based on real-world feedback.

  • Roadmap shaped by user feedback
  • Operating system and security updates
  • Ongoing feature prioritisation
02 / Technologies

Different routes. One goal: the right experience.

Native & cross-platform

Which tools should build the app?

Swift and Kotlin target a single platform. React Native and Flutter allow a shared codebase. The right choice depends on product needs and the team.

01The principle

What is a native language?

Native code is written specifically for one operating system, with direct access to its APIs and capabilities.

  • A separate codebase for iOS and Android
  • Full access to system features
  • Excellent performance, with two codebases to maintain
02Apple / iOS

Swift

Apple’s language for iOS, iPadOS and macOS development.

  • Introduced in 2014, open source since 2015
  • Modern syntax with strong, safe typing
  • Used with UIKit and SwiftUI
03Google / Android

Kotlin

The language officially recommended by Google for native Android development.

  • Created by JetBrains, endorsed by Google in 2017
  • Fully interoperable with existing Java code
  • Concise syntax with stronger protection against null errors
04One codebase, two platforms

React Native (CLI and Expo)

Meta’s framework for building iOS and Android apps in JavaScript or TypeScript from a single codebase.

  • CLI: full native configuration and control
  • Expo: quick start, managed tools and less configuration
  • Components rendered as real native elements
05A rendered interface

Flutter

Google’s framework based on Dart, with its own rendering engine.

  • One codebase for iOS, Android, web and desktop
  • Its own rendering engine for consistent results across platforms
  • Smooth animations and near-native performance
03 / Artificial intelligence

AI supports every stage.

It speeds up research, production and analysis. The team remains responsible for decisions and quality.

01

Understanding the need

Summarise interviews and feedback at scale

02

Business plan

Faster market research and simulations

03

Design

Generated wireframes and design systems

04

Planning

Refined effort estimates

05

Development

Generated code, completed and reviewed

06

Testing

Generated tests and earlier bug detection

07

Launch

Assisted store listings and compliance

08

Monitoring

Continuous analysis of issues and feedback

09

Evolution

Roadmap prioritised by actual usage

What happens next?

An app is never really finished.

Observe, learn, improve. The cycle begins again with a deeper understanding of users.

Discuss your project ↗ Back to the beginning ↑