Every time I submit an update to the App Store, I find myself wondering the same thing: what actually happens after I press the button in App Store Connect?
Developers see a handful of status messages. Waiting for Review. In Review. Ready for Distribution. Occasionally, the one nobody wants to see: rejected. Somewhere between those messages, however, an Apple employee may have opened the app, signed into it, tapped around its interface and decided whether millions of people should be allowed to download it.
It is one of the most important operations inside Apple, yet it is also one of the least visible. There is no glossy video tour of an App Review center and reviewers generally aren’t public-facing Apple employees. But years of reporting, developer documentation, regulatory disclosures and Apple’s own transparency reports give us a surprisingly detailed picture of what happens behind the curtain.
And the scale is enormous.
Apple’s 2025 App Store Transparency Report says its team reviewed 9,100,620 App Store submissions during the year. That works out at almost 25,000 submissions every day on average, although a submission is not necessarily a completely new app and the same app can be submitted more than once before it is approved.
More than 2.09 million of those submissions were rejected. Another 387,087 were eventually approved after initially being rejected.
So where do all of those apps go?
One of the most revealing accounts of life inside App Review came from CNBC in 2019. At the time, the main operation reportedly worked from two offices in Sunnyvale, California, rather than from Apple Park itself. Apple had also expanded App Review into locations including Cork, Ireland and Shanghai, China, helping the company deal with submissions from developers around the world.
The reviewers themselves were reported to be Apple employees rather than an outsourced moderation workforce. They were paid hourly, carried Apple employee badges and received company benefits.
New reviewers reportedly started with iPhone apps. As they gained experience, they could move onto more complicated submissions involving subscriptions and in-app purchases, or apps built for products such as Apple Watch and Apple TV.
Then came the part that has fascinated developers for years: the desk.
According to the 2019 reporting, a reviewer would sit at a Mac and “claim” a batch of submissions through an internal web portal known as App Claim. The app itself would then often be examined on an iPad, even when the developer had submitted an iPhone app.
That detail helps explain something developers have occasionally noticed over the years: App Review communications containing iPad screenshots of an app they thought was being reviewed for iPhone.
Was it specifically an iPad mini? I couldn’t find reliable evidence for that. Contemporary accounts describe an attached iPad but don’t consistently identify the model, so the often-repeated image of every reviewer sitting beside an iPad mini appears to go further than the evidence supports.
There were separate stations for hardware that genuinely required it, including Apple Watch and Apple TV. Apple’s current guidance also makes clear that apps dependent on unusual physical hardware can require the developer to provide a demonstration video or even the hardware necessary for review.
Once the app is running, the human review is not simply someone checking whether it launches.
Apple’s App Review Guidelines cover safety, performance, business practices, design and legal requirements. A reviewer may need to check that an app doesn’t crash, that its metadata accurately describes what it does, that purchases work correctly, that prohibited content isn’t present and that the developer isn’t attempting to hide functionality from Apple.
If an app requires an account, developers are expected to give App Review working credentials or an approved demonstration mode. Backend services must be switched on. Non-obvious features should be explained in the review notes. If two-factor authentication or a special configuration is required, Apple needs a way through it.
In other words, the reviewer is supposed to be able to get into the same app experience as a real customer.
But there is an obvious problem. There are more than 9 million submissions a year and only so many humans.
The 2019 investigation reported that reviewers were expected to process somewhere between 50 and 100 apps a day. Their throughput was tracked by internal software reportedly called Watchtower, which measured how many reviews they completed per hour. Quality mattered too: reviewers could be judged on whether their decisions were later overturned.
Those figures are now several years old and shouldn’t be assumed to describe the exact targets inside App Review in 2026. Later reporting put Apple’s reviewer workforce at nearly 500 people, but Apple does not currently publish the size of the team or an individual daily quota.
What Apple does publish shows how much the machinery around those humans has evolved.
Apple said earlier this year that App Review now combines human expertise with machine learning. Automated systems can identify suspicious patterns, compare apps for similarities and flag potentially problematic changes in updates so human reviewers can concentrate on cases that need closer attention.
Apple has also disclosed to European regulators that apps and updates go through both automated and human review before publication. Automation continues after an app is live, looking for things such as malware or “bait-and-switch” apps that change their behavior after approval. Apple says enforcement action is not taken solely because an automated system flagged an app; potentially problematic apps are escalated to human specialists.
That makes the modern review process less like a person starting from zero with every submission and more like a layered screening system, with software helping decide what deserves human attention.
Even so, humans still have a formidable workload.
Apple says 90% of submissions are now reviewed in less than 24 hours. Compare that with the operation described in 2019, when Apple’s internal service target was reportedly focused on getting 50% through review within 24 to 48 hours, and it becomes clear how aggressively the company has worked to speed things up.
The pressure around major software releases has historically been particularly intense. The 2019 investigation described long days when developers rushed to update their apps ahead of a new version of iOS. One internal email from 2018 reportedly told reviewers that 12-hour days were being opened up while the team worked through a backlog.
Again, that is a historical snapshot rather than evidence that reviewers work those hours today. But it offers a rare glimpse at what life inside the review center could be like when thousands of developers all wanted their update approved at the same time.
It also explains why your app might only spend a surprisingly short amount of time visibly “In Review.”
Historical accounts said many straightforward apps could be dealt with in only a few minutes. That doesn’t necessarily mean every feature has been exhaustively tested. The reviewer is working against Apple’s rules, the information supplied by the developer, automated checks and whatever parts of the app need to be examined to reach a decision.
More complicated apps can take longer. Apple warns that incomplete submissions, missing login information, inaccessible services or apps requiring unusual configurations can delay the process.
And sometimes an app reaches a question one reviewer can’t settle.
The 2019 reporting described an Executive Review Board that dealt with controversial or difficult decisions and helped set policy for the wider review organization. At the time, it was reported to meet weekly and was led by Phil Schiller, who has long overseen the App Store.
Apple today publicly offers developers an appeal process through the App Review Board when they believe an app has been misunderstood or treated unfairly. Developers can also communicate directly with App Review through App Store Connect and can book a 30-minute Webex appointment with the team to discuss guidelines and review issues.
Rejection itself is far from unusual.
Of the 9.1 million submissions reviewed in 2025, Apple rejected more than 2 million. Performance-related rules accounted for more than 1.35 million rejection citations, while legal requirements accounted for almost 496,000 and design rules for more than 415,000. A single submission can be cited under more than one guideline.
Apple says more than 40% of unresolved review issues are connected to guideline 2.1, App Completeness. That includes remarkably ordinary problems: crashes, placeholder content, incomplete information and apps that simply aren’t ready to be released.
There is a much darker side to the job as well.
In 2025, Apple says it rejected more than 22,000 submissions containing hidden or undocumented features, more than 371,000 that were spam, copies or otherwise misleading, and more than 443,000 for privacy violations. Nearly 59,000 apps were removed after engaging in bait-and-switch behavior, where an apparently ordinary app changed what it did after getting through review.
Reviewers therefore aren’t just deciding whether a button works or an interface follows Apple’s design rules. They sit on one of the barriers between an enormous global audience and developers deliberately trying to get malicious or deceptive software onto their devices.
That makes the anonymity surrounding App Review more understandable. Historical reporting said employees had received threatening feedback from developers angry about decisions affecting their businesses. For a developer, a rejection can mean a delayed launch or lost revenue. For the person on the other side of App Store Connect, it may simply be one decision among dozens they have to make that day.
There is also an interesting human element to the process that developers rarely see.
Apple said its review team made more than 20,000 phone calls to developers in 2022 to help diagnose and resolve issues that had caused submissions to be rejected. The company now also provides direct correspondence through App Store Connect, appeals and review appointments.
So the mysterious reviewer isn’t necessarily an anonymous gatekeeper trying to find a reason to say no. In many cases, the job is to identify why an app doesn’t meet the rules and give the developer a route toward fixing it.
Having submitted updates to the App Store myself, knowing all of this makes the little status indicator in App Store Connect feel quite different.
When it switches to “In Review,” your app may have already passed through automated systems looking for suspicious code or behavior. A reviewer somewhere in Apple’s global operation may then claim it from a queue, open it on a physical device, use the credentials and notes you’ve supplied, compare what they find against a constantly evolving rulebook and make the decision.
Approve. Reject. Or hold it for another look.
In 2025, that decision-making machine processed an average of almost 25,000 submissions every single day.
And for all of Apple’s automation, machine learning and scale, one of the most consequential moments in getting an app onto more than a billion Apple devices can still come down to a human being opening it and deciding whether it belongs there.