Gamed: Predatory Platforms and Indie Game Development

The platform economy exploits small game devs and rewards incumbents, fueling a dubious push for data analytics and constant optimization.

A comic strip panel shows an overworked game developer seeing that his sales are low on a game store platform. The next panel shows him receiving delivery of a data analytics package that guarantees success. In the last panel, the developer's game underperforms against a cheap slot-machine game.

When most game developers release a game, they hope that people will play it. Ziba Scott and Alex Schwartz had a different idea. Disillusioned by the lukewarm commercial reception of their mobile games, they noticed that seemingly popular and profitable games on the Google Play Store were largely free-to-play and monetized through advertising. Their design, such as games within the highly popular casino “slots” genre, also seemed largely low-effort and ethically dubious.

Frustrated by their lack of creative success within the rules of the system, Scott and Schwartz decided to lean into it and make it work for them. They used prebuilt video game assets, such as graphical models and textures purchased through the popular game development software Unity, to quickly create their own generic casino slot game template. With images and text pulled from Bing and Wikipedia via a script, they then generated 1,500 versions of the same game between 2013 and 2017, publishing them as free-to-play downloads on the Google Play Store.

Rather than take money from their players, the pair partnered with an ad network — a service that connects an app or game with advertisers, essentially turning the game into ad space. The more ads that people click, the more money the developer makes. Typically, a developer will want to keep players within the game so that they potentially look at (and click on) more ads. But Scott and Schwartz had another strategy. Because their games were of such low quality (“garbage,” as they later put it), they saw extraordinarily high click-through rates on interstitial ads — full-screen ads that appear at certain points throughout a game.

“We think the quality was so low in our shit,” Schwartz remarked to Ars Technica, “that the ads were a portal to a better world.”

While ad metrics like click-through rate are by all accounts pretty questionable for evaluating whether or not a person has clicked on an ad, they are the primary metric for evaluating an ad’s successful placement and for getting paid. Scott and Schwartz found that they drove their ad revenue up to $211 a day, earning $50,000 over the course of the four-year project.

Their self-described experiment in quantity over quality exposes many of the wider social and economic conditions surrounding game production, particularly in this moment of online platformization. In particular, it highlights the way that platforms embody what the sociologist David Beer calls “metric power” — that is, the ability to powerfully shape experiences through technologies and techniques of measurement and calculation, with success in the platform economy based on success according to metrics like click-through rates and app store game categories and keywords.

Scott and Schwartz highlighted how adherence to a platform’s internal logics was the chief determinant of their games’ commercial viability. As Sophie Bishop has noted in her research on vlogging as a form of platform-based creative labor (particularly on YouTube), noncompliance with platform logics results in “algorithmic invisibility,” which negatively affects a creator’s relevance and revenue.

Platformization is often framed in terms of entrepreneurial opportunity, particularly for developers who operate outside of formal working arrangements (such as a job at a big studio). You’re given a promise that you can make it on your own because you’re provided the tools to create and distribute your game. Despite this promise, the vast majority of revenues generated through mobile game sales on digital shops like the Google Play Store or Apple’s App Store accrue to US-based industry incumbents — an effect that has been described as “app imperialism.” Game developers working within the app economy largely do not earn a sustainable living from their games, and the steady influx of more developers has meant that the costs associated with game development have risen.

Despite the platform economy’s promise of providing an alternative to the video game industry of the 1980s to 2000s, when console manufacturers and software publishers were gatekeepers, it doesn’t really benefit small, independent developers. Platforms have instead become a means of consolidation for a few big-name incumbents. Though they ostensibly provide an easier pathway from development to market, they also effectively exploit creators by atomizing their work and making it “informal,” as Brendan Keogh puts it. This informalization weakens collective bargaining possibilities and employer obligations, while simultaneously generating a continuous stream of content and revenue for trillion-dollar platform companies like Apple and Google, which take 30 percent of app sales and in-app purchases. (Epic, the game developer behind the wildly popular Fortnite, filed antitrust lawsuits against Apple and Google after the game was booted from their app stores for circumventing their fees. Apple then reduced its cut to 15 percent if developers earn less than $1 million in a year, while Google decided to take 15 percent of a developer’s first million dollars in revenue before reverting to 30 percent.)

Platformization and the precarity of metrics-driven game development has fueled the emergence of an entire industry of “data analytics” — tools for the harvesting, visualization, and analysis of data. These data analytics promise to turn the art of game development into an “objective” science, providing developers with the means to efficiently factor insights around trends, competition, user activity, and so on into their work. While these tools are framed around monetizing one’s game, the appeal is by no means just to hyper-commercial freemium games, but rather to an industry of indies for whom an income stream is necessary to sustaining continued creative activity, as studies of indie developers working in Australia and Canada suggest.

Of course, data has always been a part of game development, from the data collected during quality assurance testing, to the practice of developers trawling and collecting information from social media. The capture and analysis of data through analytics has also been widely employed by larger companies, often involving bespoke tools and requiring in-house data analysts to gather, clean, and render the data actionable.

Today, analytics software programs that largely focus on optimizing user engagement and monetization are presented as off-the-shelf packages that are modular yet robust, doing away with the technical literacies and financial costs that were previously required to perform this kind of work. These tools are presented in the form of easy-to-parse dashboards with features like sparklines and bar graphs that supposedly offer developers technical insights that have typically only been available to large game studios. Independent developers are provided the means to cut out data scientists and product design teams, becoming their own data analysts.

Analytics tools are generally marketed as a solution to the systemic insecurity and risk facing independent game developers, particularly within the platform economy. “We cannot rely on our games’ quality to magically attract users,” says the analytics software company GameAnalytics on its website. Elsewhere, GameAnalytics markets itself as one of the “objective tools” that allow game developers to “estimate the risk” involved.

Developers are encouraged to depend on analytics tools to better support themselves within the precarious conditions of platform capitalism. For example, given that most mobile games are distributed for free and monetized through in-game advertising, with user acquisition increasingly achieved through (often costly) ad campaigns, tools like IronSource offer a means to “find the highest quality users for your app” by “leveraging opt-in and contextual data” and a “proprietary AI algorithm.” They offer a sense of security based on the premise of an advertising metrics system that may be completely faulty, fraudulent, and itself at risk of total collapse.

These services emphasize the scale of data to inspire confidence in its veracity. “Our analytics suite gives you key insights into gameplay, monetization, and more — letting you create the best possible experience for players,” says Unity Analytics. “1.2B players… That’s the baseline for our benchmarks and trends studies,” boasts GameAnalytics, referring to the scale of data it derives from 100,000 game subscribers. Game of Whales describes itself as “a mobile game developer’s AI tool that tracks user behavior in real-time and programmatically interacts with your players to maximize lifetime value and reduce churn” — that is, to maximize the time users spend within the game and to minimize users who leave the game.

Amazon Web Services, Amazon’s cloud computing subsidiary, has also recently become a major proponent of game analytics. AWS markets a “Game Tech” stack for data capture, storage, and analysis as an effort to democratize analytics. “You just go and click a couple of buttons and it’ll launch this whole thing as a working, viable solution,” remarked AWS architect Kyle Somers at an Amazon-sponsored panel at the 2019 Game Developers Conference.

As metrics increasingly drive workflow, data analytics can be seen as a natural response to the challenging conditions of game development. They are framed as a mode of what has been termed “careful surveillance,” where watching over something or someone is justified out of concern, wariness, or responsibility. But are these tools a solution for systemic risk and precarity? Who really stands to gain from their use?

Despite the techno-solutionist nature of many analytics tools, they are typically framed as crucial to the constant optimization of games; very few of them focus on time-saving advantages. The video game industry is already characterized by a highly individualized entrepreneurial ethos that lends itself to conditions of overwork, particularly among the independent developers that these tools are targeting. As a recent study on “games-as-a-service” by the researchers Louis-Etienne Dubois and Johanna Weststar suggests, the need to constantly develop and optimize one’s game post-launch represents a kind of “always-on” workflow — one that analytics may well bind developers to rather than liberate them from.

One of the few scholarly accounts of how game developers use data analytics underscores this point. “Mobile production for digital platforms intensifies game labor rather than facilitating its democratization in any straightforward way,” writes the sociologist Jennifer R. Whitson, who closely followed a studio that produces free-to-play mobile games. “It restricts creative autonomy, exacerbates the burden of risk on developers, and reinforces existing market and gender inequities. Rather than creatively liberating developers and expanding access to game development, data-driven design for digital platforms introduces new gatekeepers and literacies of exclusion.”

Much like many other industries undergirded by data and automation, it’s not hard to imagine various underclasses performing the “ghost work” required to support these big data tools. We already see on-demand QA testing (an area of game work that is at once highly exploitative and marginally considered) with PlayTestCloud, a service that packages its playtesting with data analytics. As analytics have become central to the infrastructures of game-making, it has become important to consider precisely who is actually involved in capturing, cleaning, and rendering the data “actionable.”

More generally, the promise of data analytics services to mitigate the high-risk nature of game development undermines the idea of game development as a communal practice while obscuring the potential for more radical approaches to workplace autonomy. As Whitson argues, rather than relieve today’s pressures of entrepreneurship in the games industry, they individualize risk and intensify metric power as well as many of the systemic inequities that characterize game development. The platform companies and analytics providers benefit the most, with analytics services generating revenue from subscriptions and the accumulation of data that enable them to act as brokers in the broader market and offer costly “market insights.”

This is not to say that new, constructive arrangements of developers and technology aren’t possible. Some critics of the state of game development advocate free open-source tools like Twine or Bitsy, which support platform alternatives that are more collective and cooperative. Such low-tech approaches are both aesthetically and politically distinctive, with rougher edges taking the place of ongoing optimization and intensive work. As the video game ethnographer Aleena Chia and her colleagues have noted, the game-making network effects of software like Twine have been repurposed in the service of “a fringe of otherwise marginalised videogame developers,” who “consequently produced a range of games that directly contravened accepted videogame conventions.”

Open game marketplaces like, particularly those that support releases made on alternative game engines, further legitimize different platform approaches. For example, in contrast to the visibility advantage the typical app store gives popular incumbents, visibility might be sorted to favor new games rather than popular games. Beyond new digital tools, not-for-profit cooperatives are also fostering sustainable practices for independent game development and distribution. 

Perhaps in the future we might see new platforms emerge to meet the needs of these organizations and build equitable systems that can enable game developers to operate on their own terms. After all, to subvert metric power, we need to make the metrics work for us.

Follow The Reboot

Join a growing community that’s examining the state of the internet and exploring its future. Subscribe to our newsletter.

A comic strip panel shows an overworked game developer seeing that his sales are low on a game store platform. The next panel shows him receiving delivery of a data analytics package that guarantees success. In the last panel, the developer's game underperforms against a cheap slot-machine game.

Artwork By


Contact Us

Have an idea for a story or illustration? Interested in discussing partnerships? We want to hear from you. Send us a note at info(at)thereboot(dot)com.

Recommended Reading