When Airbnb killed React Native in 2018 — the single most over-quoted moment in this entire debate — the numbers buried in their own post-mortem told the story better than the headline did. At the peak of its bet, Airbnb had around 220 screens running in React Native and roughly 880 native screens per platform behind them. Even the famous React Native shop was about 80% native. That is the fact to hold in your head before you go shopping for a "React Native consultant," because almost everyone selling that title is quietly pretending the other 80% doesn't exist.
The conventional pitch is by now a reflex. You want a mobile app, someone tells you to hire a React Native consultant, and the brochure writes itself: one codebase, one team, ship to iOS and Android at once, half the cost of going native twice. It is framed as a procurement decision — find a cheaper way to get two apps. That framing is the problem. It treats the framework as the product you're buying, when the framework is the part that was never going to be hard.
The label selects for the wrong person
"React Native consultant" is a search term that, more often than not, surfaces a React web developer who added the word "Native" to a LinkedIn headline and never opened Xcode in anger. That is not a cheap shot; it is the predictable result of how the market is structured. React Native is JavaScript and React, which millions of web developers already know. Telling a web developer they can now "do mobile" without learning Swift or Kotlin is the most attractive lie in the field, and the title launders it. The buyer hears "mobile expertise." The seller heard "I don't have to learn the platform."
You can find the tell in Airbnb's own debrief and in nearly every honest write-up since: engineers ended up maintaining three environments — the JavaScript layer, plus iOS, plus Android — and when a feature could be written either way, people quietly chose the platform they were more comfortable with, which produced worse code, not faster delivery. The consultants who can only live in the JavaScript layer are exactly the ones who can't make that call. They don't choose React Native for the right reasons; they choose it because the alternative is a language they don't speak.
"Write once, run everywhere" was always the brochure, not the build
Here is where the consensus over-corrects in the other direction. People wave the Airbnb exit around as proof that React Native is dead, and that reading is just as lazy as the brochure it's rebutting. The same Airbnb survey found that around 63% of their engineers would choose React Native again and about 74% would consider it for a new project. React Native didn't fail them. The bolt-it-onto-a-mature-native-app strategy failed, and the inability to hire people who were fluent in both worlds failed. Walmart, by contrast, moved to React Native from a hybrid web-view app and was happy with the graphics performance and resource usage; their complaint was debugging, not viability. The lesson across both is not "React Native good" or "React Native bad." It is that the outcome was decided by who was holding it — which is a hiring question, which is the consultant question.
The framework is the easy 80%. You're paying for the 20%.
A competent developer can get a React Native screen rendering on both platforms in an afternoon. Lists, forms, navigation, a settings page — the framework genuinely is good at this now, and the modern architecture that replaced the old asynchronous bridge with a direct JavaScript-to-native interface closed much of the performance gap that haunted 2018. None of that is what you hire a consultant for. You hire for the 20% where React Native runs out of road: a native module that doesn't exist yet and has to be written in Swift or Kotlin; a gesture or animation that has to feel right on a platform with its own conventions; a third-party library that supports iOS well and Android poorly and leaves you to make up the difference; an App Store rejection; a frame-rate cliff on the cheap Android handset your actual users own. A consultant who can't open Android Studio when that day comes is not a mobile consultant. They are a React developer who will bill you by the hour while they read Stack Overflow.
Why agencies love selling you "React Native"
Follow the staffing economics and the marketing makes sense. React Native lets an agency put cheaper, more plentiful web developers on a mobile project and call the result mobile expertise. One advertised skill, a deeper bench to draw from, a lower blended rate, a tidy margin. That is also why so many of these engagements quietly become body-shopping arrangements where the faces rotate and nobody owns the hard native edges. The economics reward selling the framework as the whole job. The framework is not the whole job.
What a real one looks like
The honest version of this work is deeper, not wider. At EltexSoft our mobile group is built that way on purpose: native iOS in Swift, SwiftUI and UIKit, native Android in Kotlin and Jetpack Compose, and React Native — not three competing camps, but engineers who can pick the right one and drop to native when an app earns it. That is the whole point. The shipped work that we're proud of includes an app featured at Apple's WWDC and a coupon app that hit #2 in its App Store category and roughly two million installs — neither of which you get from people who treat the native layer as someone else's problem. The same discipline that makes this real is unglamorous: DevOps owned by the engineers who write the application code, and every pull request reviewed by another senior before it merges. The signal you want from a consultant is not enthusiasm for React Native. It is the willingness to tell you when not to use it.
The cost math everyone gets backwards
The savings are real — when the app fits. A genuine cross-platform build can be meaningfully cheaper than maintaining two native codebases, and for a content-driven, conventional app that is a sound, even obvious choice. The trap is the inversion: when an app that needed native gets forced into React Native by someone who couldn't do anything else, the "savings" turn into a rewrite, and a rewrite is the single most expensive thing in software. At our rates — roughly $50–99 an hour, a four-person pod landing around $25,000–$55,000 a month — the difference between a consultant who scopes the fit honestly and one who defaults to the only tool they own is not a line item. It is the difference between shipping and starting over. The cheaper hourly rate that can't tell those two situations apart is the most expensive decision in the project.
If the model itself is still fuzzy, start with our full breakdown of when staff augmentation is worth it. And when you want to see how we structure engagements around retention, our staffing services lay it out.
So stop shopping for the title
Do not hire a "React Native consultant." Hire a mobile engineer who happens to use React Native and will argue with you about whether you should. Bring them the right opening question — not "should I use React Native?" but "show me where React Native wasn't enough and what you did about it." If the answer involves Swift or Kotlin, a custom native module, a profiler, and a story about a platform-specific bug they actually fixed, you've found the person. If the answer is a portfolio of screens and a speech about velocity, you've found a web developer with a costume. The framework was never the hard part. The judgment about when to leave it is the entire job, and that is the only thing worth paying a consultant for.