Platform Papers is a blog about platform competition and Big Tech. The blog is linked to platformpapers.com, an online repository that collects and organizes academic research on platform competition.
By Ben Leyden.
Imagine you are a software developer who spent months refining your new mobile application, releasing it to enthusiastic users, and carefully building a strong reputation via the platform’s rating system. Each positive review lifts your visibility, bringing new users and driving further success. After a few months, you want to implement new features that your users are asking for or fix that pesky bug some users are facing, but you know that every time you enhance your app, all your hard-earned stars will vanish overnight. Suddenly, the system intended to help users make informed decisions and to bring attention to innovative and high-quality products inadvertently punishes innovation. In a study recently published in the International Journal of Industrial Organization, I investigate this exact scenario, exploring how the design of the rating system for Apple’s iOS App Store unintentionally discouraged developers from improving their products.
Why Ratings Matter for Digital Platforms
Rating systems are crucial for digital marketplaces. They provide critical information, reduce uncertainty for potential buyers, and help consumers navigate the vast array of products available online. However, the ostensibly straightforward task of summarizing individual ratings often involves significant complexity, particularly when consumers differ in how they review products, when reviewers are influenced by existing ratings or wish to avoid harming market participants, and when the quality of products changes over time. Moreover, product ratings have been shown to play an influential role in driving demand. Ultimately, platforms' decisions about how these ratings are aggregated and displayed have the potential to dramatically shape market outcomes.
How Rating Aggregation Influences Innovation
For nearly a decade, Apple’s App Store reset an app’s average rating each time its developer released an update. The rationale behind this rating reset policy might initially seem reasonable—it encouraged developers to produce high-quality updates by making ratings version-specific. Moreover, this approach offered forgiveness to developers who initially released low-quality products—they were always just one impressive update away from redemption.
However, this approach was deeply unpopular with developers. As one developer put it in 2014:
“With App Store ratings getting reset on every minor version, we are actively discouraged from updating our apps once they have good ratings. I've just submitted a new version of Delicious Library 3, and I'm scared out of my head that the first two or three people who review it won't like it, which will tank it to the extent where nobody discovers it anymore, so there won't be any positive reviews to balance them out.”
Indeed, Apple’s policy produced perverse incentives. Apple’s rating reset policy meant that high-end developers faced a reputational cost for updating in addition to regular development costs. For some games and apps, that cost might be limited—Angry Birds or Microsoft Word might rebuild their average rating within minutes. For others, however, it could be long-lasting—a high-quality but niche app might take weeks to rebuild its previously sterling reputation. Similarly, this policy had the potential to inefficiently distort development at the low end—low-quality developers could escape their bad reputation, at least temporarily, by releasing an update.
In mid-2017, Apple unexpectedly sided with developers on this issue, with the executive in charge of the App Store explaining:
“…some developers don't like submitting their updates because it resets the rating. So, they would get upset, saying, ‘Oh, man, I have a choice: fix some bugs and blow away my ratings or keep my high rating. I have a 4.7; I don't want to submit it.’ And I thought that was kind of stupid.”
Under the new policy, developers faced a choice every time they updated their products: reset their rating (as if under the original system) or maintain and build on their past reputation.
Understanding the Effects of the Rating Reset Policy
I leverage Apple’s unexpected policy change to estimate how the rating reset policy affected developer efforts. By examining extensive data on app ratings and updates, I document significant shifts in development strategies triggered by the policy change.
Developers' responses to the removal of the reset policy were immediate and sustained. They increased the frequency of product updates by roughly 35%, moving from averaging an update every seven weeks to roughly every five-and-a-half weeks. Additionally, by analyzing update “release notes,” which describe the content of product updates, I find that the reset policy decreased the frequency of both feature-adding and bug-fixing updates.
Perhaps the most striking finding is the disproportionate adverse impact the reset policy had on higher-quality apps. These apps, typically benefiting the most from positive ratings, faced substantial reputation costs every time they considered an update. My analysis shows that highly rated app developers faced the strongest incentives to avoid rating resets. Consequently, precisely the apps consumers might desire most—those with ongoing improvements, frequent innovations, and responsiveness to user feedback—were systematically discouraged from innovating due to fears of rating loss.
Did Fewer Product Updates Translate to Lower Quality?
My analysis finds that developers released fewer updates in the presence of Apple’s original rating reset policy. This could reflect a reduction in innovative effort by developers, which would be bad for users (and, therefore, the platform) who benefit from frequent improvements in app functionality, security enhancements, and the introduction of user-driven features. Alternatively, developers might have strategically bundled updates into fewer, larger releases. While the latter approach—reflecting a suboptimal release schedule for new features—would still hurt users and the platform, it would be significantly less costly to society.
To better understand the quality impact of the reset policy, I analyze how average ratings and the number of new ratings and reviews, a proxy for new user demand, were affected by the policy change. I find that both average quality and the number of new ratings increased after the policy change. Thus, the rating reset policy paradoxically undermined the App Store’s overall quality, limiting consumer access to continuously improving digital products.
About the research: For nearly a decade, Apple's iOS App Store reset developers' salient user rating scores with every app update, potentially discouraging high-quality apps from updating. Using two years of app-level data from the App Store, I find that when Apple removed its reset policy in mid-2017, product updating increased by 35%. This effect was immediate and sustained. Highly rated apps were most affected by the policy change, as were apps whose repetitional penalty from the reset was likely to be long-lived. Apple's rating reset policy had a material, adverse effect on the quality of products on its digital platform, as both average app quality and a proxy for consumer demand increased following the policy change.
Managerial Implications
My findings underscore the importance of deeply understanding how platform policies affect strategic incentives for firms operating in digital markets. Platforms must carefully evaluate the broader implications of their policies and systems governing conduct on their platform, including, but not limited to, rating systems, and ensure that mechanisms intended to encourage high-quality content do not inadvertently create perverse incentives. Platforms that seek sustained competition and innovation, and, ultimately, user satisfaction must align their policies and platform design more closely with these goals. The lessons are equally vital for firms operating on platforms, who must consider how product development, pricing, and related strategic decisions interact with underlying platform mechanics.
Platforms hold enormous power—not only in shaping consumer choices but also in directing the trajectory of intra-platform competition and innovation. My research into Apple’s App Store rating system provides clear evidence that, if poorly designed, these systems can create significant barriers to innovation, ultimately harming users, developers, and the platform itself.
This blog is based on the author’s research, which is published in the International Journal of Industrial Organization and is included in the Platform Papers references dashboard:
Leyden, B. T. (2025). Platform design and innovation incentives: Evidence from the product rating system on Apple's App Store. International Journal of Industrial Organization, 99, 103133.
Platform Updates
The Platform Papers references dashboard saw one of the biggest monthly updates in newly added papers to date! 22 published academic studies on platform competition were added in June, bringing the total to 784 articles.
WhatsApp opens the door to ads - Meta will now show sponsored content in WhatsApp’s Stories-like Status feature, placing ads alongside updates from friends and family. The long-anticipated move brings targeted advertising to the platform, as Meta seeks new revenue streams without interrupting personal chats.
Apple hits the brakes on boldness at WWDC - This year’s keynote saw Apple opting for safety over spectacle, highlighting a new Liquid Glass look and tighter device integration. Skipping AI hype, it doubles down on its hardware-software edge and developer focus, favoring present-day polish over dramatic promises.
Apple’s AI ambition stuck in neutral - A year after unveiling “Apple Intelligence,” Cupertino’s AI features remain delayed and underwhelming, with rivals like Google and OpenAI racing ahead. Its latest releases center on productivity tweaks and design changes, leaving the company searching for its next big vision.
Meta stokes AI talent arms race - OpenAI’s Sam Altman reveals Meta is offering $100M bonuses to lure AI experts, supercharging the competition for top minds. The big bets come as Meta launches a $15B super-intelligence push and invests in Scale AI to fuel its next wave of AI breakthroughs.
Nintendo Switch 2 smashes launch records - The new console moves 3.5 million units in just four days—the strongest debut in Nintendo’s history. With upgraded hardware, magnetic Joy-Con 2 controllers, and innovative GameChat features, Switch 2 is set to redefine how fans race, play, and connect.
Amazon’s Prime delivery hits rural America - The e-commerce giant is rolling out same-day and next-day delivery to 4,000+ small cities and towns in the US, aiming to boost demand beyond big metro markets. Amazon is investing $4B to capture more frequent purchases from underserved regions.
GENIUS Act sets stage for stablecoin surge - New Senate bill hands Treasury broad oversight and clears the way for banks, fintechs, and retail giants like Amazon and Walmart to enter the dollar-backed crypto race. Payment networks brace for shakeup as the crypto industry scores a major win in Washington.
Tesla’s robotaxis make bold Austin debut - Elon Musk rolls out Tesla’s first driverless cabs for paying passengers in Austin, marking a major step toward autonomous ambitions. Regulatory challenges and fierce competition from Waymo loom large, while Texas also sets new rules for autonomous vehicles.
Platform Papers is published, curated and maintained by Joost Rietveld.