Mobile development has a reputation for being slow, complex, and harder than it needs to be. Platform quirks, rigid review gates, and ever-growing app complexity can make it feel like the toolchain is working against you.
But the data tells a different story. We analyzed tens of millions of builds across thousands of mobile teams on Bitrise, spanning three years of real-world data from 2022 to 2025. The results challenge some common assumptions, and confirm others.
We sat down with Arpad Kun, VP of Engineering at Bitrise, to unpack the findings.
Q: The Bitrise Insights Report shows build times dropped 28% while workflows got more complex. How is that possible?
Arpad: This was one of the clearest signals in the data. Teams aren't just building faster, they're doing more per build. The average number of steps per workflow increased by 23% over the same period. Teams are adding security scanning, extra test stages, and code signing verification as real quality gates, and still coming out ahead on speed.
The drivers are a combination of faster computer hardware, smarter build caching, and better pipeline architecture. On caching specifically, our data shows a median cache hit rate of 57% on primary workflows, with top-performing teams hitting nearly 90%. That translates to a roughly 25% reduction in build time at the median, and 30% for the slowest builds.
The takeaway is that speed and quality aren't a trade-off. The best teams have figured out how to get both.
Q: Why does mobile build speed still matter so much?
Arpad: Speed is really about feedback loops. It's not just "fast builds good, slow builds bad," it's about how quickly your team can learn from what just happened and act on it.
When a build takes 30 minutes, your engineer context-switches, picks up something else, and by the time the result comes back, they've lost the mental model. When it takes 9 minutes, which is our current median, they're still in the zone.
This matters even more as AI becomes part of the development workflow. AI-assisted coding and automated optimization depend on rapid iteration cycles. The tighter your feedback loop, the more effective those tools become. Slow pipelines become a bottleneck not just for humans, but for any automated process trying to iterate on your code.
Q: One of the most surprising findings is that flaky tests are getting worse, not better. What’s behind that?
Arpad: This one stood out. In 2022, the probability of a team encountering a flaky test in their runs was around 9%. That number has gone up every year since.
Now, an important caveat: our customer base has shifted toward larger enterprise teams running more complex test suites. More tests inherently mean more surface area for flakiness. So part of this trend is compositional, but it's still a real problem that teams are dealing with daily.
Flaky tests in mobile are more than three times more likely than in web testing. Network conditions, device state, OS-level timing behaviors, mobile has a lot of environmental variables that are hard to control.
Q: What makes flaky tests so damaging?
Arpad: They cost you in three ways.
Time: every flaky failure triggers a rerun. Engineers have to investigate, confirm it's not a real issue, and restart the pipeline. In many cases, nothing was actually broken, the test just failed for no reason.
Money: every rerun burns compute. It's wasted infrastructure spend with zero value.
Trust: this is the worst one. When tests are unreliable, engineers stop believing in the results. They start ignoring failures, skipping test stages, or pushing through despite red signals. Once you lose trust in your test suite, you've effectively lost the safety net.
Q: The data shows that teams with better pipeline visibility perform significantly better. Why is this?
Arpad: Because you can't fix what you can't see. It sounds obvious, but most teams don't have good visibility into where their time actually goes.
Our data shows that teams actively using pipeline monitoring, specifically Bitrise Insights, consistently achieve higher build success rates year over year. When you can see which tests are flaky, which steps are slow, and where failures cluster, you can prioritize fixes based on actual impact instead of guessing.
As pipelines get more complex, and they are, with 42% of builds now running in multi-workflow pipelines, up from 16% in 2023, this visibility becomes essential. You can't manage a pipeline with 5-11 parallel workflows by gut feel.
Q: React Native more than doubled its build share in the report. What's driving this increase?
Arpad: React Native went from about 11% of all builds in 2022 to nearly 26% in 2025. That's a massive shift. And it's not just a few large teams skewing the numbers, when we look at it by customer count and repository count, the growth is broad-based.
What we're seeing is organizations consolidating around cross-platform frameworks to reduce duplication and share code across iOS and Android. React Native specifically benefits from the enormous JavaScript/TypeScript talent pool and a mature ecosystem. Flutter remains steady at around 4-5% but hasn't seen the same acceleration.
The practical impact is that cross-platform builds now account for nearly a third of all builds on our platform, up from about 17% in 2022. That's a fundamental shift in how mobile teams are structured and how they think about their codebase.
Q: Xcode adoption is another area where the data reveals real tension. Can you talk us through that?
Arpad: When Apple releases a new major Xcode version, the highest-volume teams adopt it within about four weeks. But it takes the majority of teams 19 to 21 weeks, nearly five months, to make the switch.
That gap matters because Apple increasingly enforces adoption deadlines. We've seen situations where Apple released a version, found a critical issue, pushed a fix, and within a week required teams to rebuild with the new version just to keep submitting to the App Store.
If your team takes months to adopt, you're not just behind on features, you risk being unable to ship at all. In regulated industries like financial services, where app updates often have compliance implications, even a few weeks' delay can be costly.
That's why we make new Xcode versions available within 24 hours of Apple's release. The infrastructure shouldn't be the bottleneck.
Q: What are the top things engineering leaders should take away from the Bitrise Insights report?
Arpad: It comes down to three key things.
First, speed and quality aren't a trade-off, the data proves it. The best teams are both faster and more thorough. Invest in caching, invest in faster hardware, invest in modular pipelines.
Second, flaky tests are a bigger problem than most teams realize, and they're getting worse. Don't accept flakiness as normal. Instrument your test suite, track flake rates, and treat persistent flakers as bugs with real cost.
Third, visibility is the foundation for everything else. You can't optimize what you can't measure. Teams with good observability across their pipeline consistently outperform those flying blind.
The mobile ecosystem is only getting more complex, more frameworks, more quality gates, more compliance requirements, faster release expectations. The teams that invest in their foundations now are the ones that will be able to absorb that complexity without slowing down.
Check out the full report
Want to explore all the data behind these findings? Read the full Bitrise Mobile Insights Report.
