In this article
Nobody can price a React Native app from the question alone, and anyone who does is quoting a number they will revise later. But "it depends" is not a useful answer either. This article explains how we estimate, what moves the number, and the ranges we see across the 25 projects in our case studies.
Cost is people multiplied by weeks
Every estimate we produce is built from two inputs: how many people, for how many weeks. A first production version of a React Native app usually takes a small senior team between 12 and 20 weeks. That team is typically two engineers, a part-time designer, a dedicated QA engineer and a project lead. Multiply the weeks by the team's blended weekly rate and you have the budget. The rest of this article is about what changes the weeks.
We prefer this framing because feature lists are misleading. "Login" can take a day or a week depending on whether it means email and password, or social sign-in plus SSO plus biometric unlock plus account recovery. Counting weeks forces the conversation onto what each feature has to do, and for whom.
What moves the number
Backend and integrations
The app on the phone is often half the project. If it needs its own API, admin panel, database and hosting, add a backend engineer for most of the timeline. If it has to talk to an ERP, a payment gateway, a CRM or a hospital system, each integration is a separate line, and legacy integrations are usually the longest one. We scope them first because they are where estimates go wrong.
Offline use
Office apps can assume an internet connection. Field apps cannot. Local-first storage, a persistent sync queue and a conflict policy add real engineering, and the testing is heavier, because the conditions that break sync are weak and intermittent connections rather than airplane mode. Our field service compliance platform and offline-first inspection app both carried this cost, and both would have failed in the field without it. We describe the architecture in a separate article.
Hardware access
Camera, GPS and push notifications are routine. Bluetooth Low Energy, background location, NFC and anything that talks to a physical device are not. They need native modules, testing on specific devices, and time for platform quirks that only show up on real hardware. Plan for extra weeks and a drawer of test phones.
Real-time features
Chat, live bidding, dispatch boards and location tracking need a real-time layer, usually WebSockets or a managed service, plus the state handling to keep the screen correct when updates arrive out of order. These features look small on a wireframe and are not.
AI features
An assistant, document intelligence or a voice interface adds model integration, retrieval, evaluation and guardrails. The integration work is manageable. The testing and iteration take longer, because the output is probabilistic and the first version is never the one that ships. Our AI knowledge assistant app is a representative example.
Compliance and review
Health, finance and anything involving minors adds review cycles: legal, clinical stakeholders and app store scrutiny of health content. These are calendar weeks more than engineering weeks, and they belong in the plan from the start.
Design readiness
If you arrive with finished designs, we start building in week one. If you arrive with an idea, we start with one to two weeks of discovery and design. The second path costs more up front and less overall, because designing during development produces rework.
Typical ranges by app type
These are indicative team-weeks for a first production release with a small senior team, drawn from projects we have shipped. Yours will differ, and discovery exists to find out how.
| App type | First version | Main cost drivers |
|---|---|---|
| Content or community app | 10 to 14 weeks | Feeds, profiles, media, moderation |
| Consumer app with payments | 12 to 16 weeks | Gateway integration, subscriptions, store compliance |
| Field service or inspection app | 16 to 22 weeks | Offline sync, photo evidence, multi-role workflows |
| Marketplace app | 18 to 26 weeks | Two-sided onboarding, split payouts, messaging, disputes |
| Bluetooth or IoT-connected app | 16 to 24 weeks | Native modules, device testing, firmware coordination |
| AI-assisted app | 14 to 22 weeks | Model integration, retrieval, evaluation, guardrails |
Multiply by the team's weekly cost to get a budget. A 12-week consumer app and a 24-week marketplace app end up at very different totals, and the difference is almost entirely in transaction logic and integrations rather than in the number of screens.
Where estimates go wrong
- Estimating from a feature list without asking what each feature must do. This causes most surprises and is avoidable in a two-hour conversation.
- Forgetting the backend. The app is visible; the API, admin panel and hosting are not, and they are often half the work.
- Treating launch as the finish line. Operating systems update, store rules change and dependencies age. A realistic plan includes maintenance from the first month.
- Taking the cheapest quote. We have inherited enough React Native codebases to know that the cheapest first version is often the most expensive one over two years. If you are already in that position, our React Native practice covers codebase rescue.
Getting a number you can rely on
Send us a feature list, the app you want to replace, or a competitor you admire with notes on what you would change. We run a free scoping exercise and come back with a team, a timeline and a staged budget. The engagement models page explains how fixed-price and dedicated-team arrangements differ, and the first conversation costs nothing.