For this exercise, I proposed a taxonomy of innovations and transformations within the mobile free-to-play category that are hierarchically ranked based on breadth of impact:
- Evolutions are market-level changes — such as the introduction of the smartphone or the Games-As-A-Service model — that impact consumer behaviors and fundamental game design principles;
- Revolutions are business model changes that inform the ways that free-to-play games are monetized, such as the adoption of in-game advertising, subscription packages, and battle passes;
- Permutations are game design changes that impact game-level interaction and engagement behaviors.
My argument in the aforementioned piece is that ATT fundamentally alters the market for free-to-play mobile games: it impacts not just game design choices (permutation) or the ways in which monetization is implemented (revolution), but the fundamental character of the free-to-play gaming business model across distribution, monetization, and engagement.
The consolidation that I predicted in my 2020 piece has mostly materialized: the dollar volume of gaming M&A transactions in 2021 was nearly 3 times that which took place in 2020. And 2022 saw more dollar deal volume still, with the two largest video game acquisitions in history announced in quick succession: Take Two’s acquisition of Zynga and Microsoft’s acquisition of Activision Blizzard, which owns King, the developer behind the perennially chart-topping Candy Crush franchise. Consolidation within the free-to-play mobile gaming category has already been realized and is indeed at least a contributing catalyst to the upheaval that is playing out, broadly, across the category.
In this article, I’ll provide an overview of how I believe the mobile free-to-play gaming category will continue to transform in the medium term now that ATT is better understood, and what developers should be cognizant of as they navigate that evolution.
Free-to-play as a business model did not precipitate the mass proliferation of smartphones — it arose as a reaction to it, and consumers have propelled free-to-play to total ubiquity on smartphones with their wallets. While SensorTower estimates that consumer spend within the mobile gaming category will have declined by a little more than 2% in 2022, it forecasts a return to modest CAGR from 2023. Free-to-play is a genie that can’t be put back in the bottle: once consumers become acclimated to a price point of $0, increasing the cost of access is a fight against economic gravity. Operating challenges won’t cause the extinction of free-to-play. In this piece, I’ll attempt to map out the changes I see taking root within the free-to-play market, with color on how those changes should inform developer strategy.
Has ATT actually impacted the mobile free-to-play market?
The appropriate starting point in ascertaining the post-ATT future of mobile free-to-play gaming is considering whether ATT has systemically impacted the mobile free-to-play gaming market in the first place. I proposed in IDFA deprecation: winners and losers, written in December 2020, that the various segments of the mobile gaming market would experience differing degrees of impact from ATT: notably, I posited that high IAP-monetization games, such as those in the social casino category, would see an extreme magnitude of impact, versus a minimal overall impact for moderate IAP-monetization games, such as those within the casual category.
This spectrum of impact seems to take shape in Playtika’s latest earnings report. Playtika is a scaled, diversified, multi-studio mobile gaming company: it operates a legacy portfolio of social casino games and a portfolio of casual games, many of which were developed by studios that the company has acquired. While revenue for the quarter beat analyst expectations, it was mostly flat on a year-over-year basis at 1.9% growth, and that growth was delivered from the company’s portfolio of casual titles. From the earnings call:
Revenue across our Casual games grew 14.4% versus a year-ago…Casino themed games revenue for the third quarter was down 10.2% versus a year-ago…The digital user acquisition environment continues to evolve and costs per install have increased in the third quarter. As we look out to our plans for 2023, we will continue to increase marketing investment in our growth franchises while being disciplined and data-driven in how we allocate marketing capital.
One question that has reverberated across the consumer tech space recently is whether macroeconomic forces — which is to say, perilously high and persistent inflation, especially in Europe — are actually weakening consumer demand and thus depressing the market for digital advertising. In the early weeks of COVID I wrote a post titled What happens to free-to-play mobile gaming during a recession? in which I supposed that a recession would be detrimental to the free-to-play gaming market because in-app purchases can be classified as a luxury good, and luxury goods generally experience demand destruction during a recession. From the piece:
But what about mobile games? Mobile games are mostly free: they are the ultimate discount entertainment. But my belief is that, while free-to-play games may also serve as a substitute good — to all forms of entertainment, including traditional console video games — the digital products purchased via IAPs in free-to-play games need to be classified independently from the games in which they exist. The decision a consumer makes to download a free game is totally separate from the decision they make to purchase a digital good in that game. Putting aside design factors (“pay to win,” etc.), it’s tautologically true that free-to-play games can be played for free, and so IAPs have to be seen as luxury goods: characterized by desire and convenience, with demand tracking income.
The CEO of Take Two, another public gaming company that operates a cross-platform portfolio and which recently completed its acquisition of mobile gaming developer Zynga, made a similar point in the company’s most recent earnings call:
There’s a difference between user acquisition and retention and conversion. And on user acquisition, I think we and everyone else is just going to be a bit more selective to drive efficiency. And in terms of retention and conversion, that’s a reflection of people just not having to spend mobile. You can enjoy a title without spending in-game. 90-plus percent of consumers do not spend in-game. And so at times when consumers are feeling the pressure of higher prices for fuel and food, for example, they may be less likely to spend money on entertainment, especially when you can have the experience anyhow. So in terms of our expectations, I think this is — it’s a moment where we will, in fact, tune up our UA spend to become more efficient. That’s a good thing.
As I state in the three-part Mobile Marketing Winter series, the current turmoil facing mobile advertisers — but especially mobile games advertisers — can mostly be attributed to three factors:
- A weak and potentially worsening macroeconomic environment globally;
- A reversion to pre-COVID engagement norms for consumer technology products;
Parsing apart the individual impacts of these factors from the composite is impossible, and the individual impacts are largely beside the point; the confluence of all three at the same time underscores why this particular moment is so challenging for mobile advertisers (and mobile advertising platforms). But I’ve argued that certain ad platforms (namely: the ones facing extreme impact from ATT in the above-linked article about winners and losers) are exaggerating the impact of “macro” factors, because those macro factors will at some point subside, whereas ATT is systemic and permanent.
Thus, the question of ATT’s burden on mobile games returns to the question of why Playtika would see contrasting performance between its social casino and casual portfolio. If macro forces are creating a drag on the mobile free-to-play gaming market, wouldn’t that drag apply universally and not selectively? I’d contend that the bifurcation in performance across Playtika’s portfolio is explained by ATT and not macro forces, which is also supported by the fact that the casual portion of Playtika’s portfolio actually grew impressively by 14%.
Teasing apart the individual effects of the three factors described above is difficult if not impossible, but it’s also only really even relevant if one believes that an actual recessionary economy could further degrade operating conditions within the mobile gaming market in the future. Unfortunately, the consolidation described above limits the data that can be used to even make an attempt at this exercise: very few pure-play, publicly-listed mobile gaming companies exist at this point, with the recent acquisitions of Glu by EA, Zynga by Take Two, and Next Games by Netflix. Looking at the quarterly revenue performance for Applovin’s games portfolio and Playtika depicts a COVID surge and a subsequent decline:
But demarcating the introduction of ATT muddies the picture. The peaks for both of these revenue curves seem to coincide fairly tightly with Apple’s rollout of ATT, which reached a majority scale of iOS devices in Q3 2022:
Again: there’s no way to dissociate the combined effects of the three factors I outline in the Mobile Marketing Winter series. But I believe — and have made the case, repeatedly — that ATT has established a permanent headwind on all businesses dependent on mobile advertising for growth. It seems reasonable, if not simply prudent, to assume that ATT is responsible for some significant proportion of the impairment levied upon these businesses and to set course accordingly.
Below, I’ll outline three fundamental changes that I believe will take shape within the mobile free-to-play gaming ecosystem and explain how developers can accommodate those changes in their operating models.
Supremacy of centralized publishing services
One of the powerful commercial innovations that the app store model instantiated was giving developers direct access to publishing technologies that could reach millions — now billions — of consumers. This was truly a marvel in 2008, when the App Store first launched: developers could build a software product and distribute it, at no marginal cost, through a digital storefront without the need to work with a publisher tasked with negotiating distribution agreements with retail outlets.
Of course, this model incurred frictions from almost the very beginning, especially as the freemium model rose to prominence and free-to-play gaming became ubiquitous on smartphones. “Access to consumers” is not distribution. The companies today that are the largest providers of fundamental marketing infrastructure for mobile game developers — companies like Applovin (2012), Unity Ads (Applifier, 2008), ironSource (2010), Appsflyer (2011), etc. — were all roughly founded just as the free-to-play model was beginning to gain traction on mobile. I don’t see it as a coincidence that what I’ve called the “2012 vintage” of successful, top-charting games, some of which still adorn the Top Downloaded chart to this day, were launched at the same time as the infrastructure needed to distribute them.
I spoke of the winner-takes-all nature of the app economy as early as 2014: as smartphone penetration increased, so did the size of the free-to-play mobile gaming market, and competition for that market was reflected in the cost and complexity of marketing games through direct response user acquisition through the aforementioned channels — and, critically, Facebook (now Meta) and Google.
User acquisition economics in free-to-play are driven by the auction systems on which the various direct response ad channels rely. I provide an overview of the different classes of ad channels for mobile app install advertising campaigns across Self-Attributed Networks (SANs) and “Broker Networks” in this Quantmar thread, and this Mobile Dev Memo podcast episode walks through the mechanics of advertising auctions. But essentially: game monetization determines the cost-per-install price a developer is able to pay for installs for various audience segments.
As Google and Meta improved their ability to target individual players on the basis of specific behavioral patterns, those companies took enormous share of the market for app installs and unlocked an incredible amount of value for free-to-play game developers, which were able to find relevant audiences fairly frictionlessly through their self-service ad platforms. Monetization and mass appeal dictated user economics, and developers who could either build robust in-game economies, appeal to very large and diverse audiences, or both, could efficiently scale their games through user acquisition efforts. While these efforts required infrastructure, it was possible for developers to scale investment into that infrastructure with ad spend: a small studio could launch and gain traction with off-the-shelf attribution and analytics tools through the power of these direct response advertising tools. Proving product-market fit and exhibiting the potential to generate meaningful revenue could be accomplished with a small marketing team and little or no proprietary advertising infrastructure.
This changes with ATT. Not only has the distribution power of the Self-Attributing Networks like Meta, Google, and Snap been diminished through the disruption of user-level behavioral profiling, but the Broker Ad networks also face attribution frictions because of the inprecision of SKAdNetwork (which will be exacerbated when or if device fingerprinting is policed by Apple). Scaling a game to initial traction is no longer as simple as turning on app advertising spend on Meta, Google, and other channels and relying on various SaaS tools to provide insight into efficiency and profitability. Advertising performance measurement now requires probabilistic tools that not only mostly don’t currently exist as SaaS solutions but might not be able to exist as SaaS solutions given the bespoke nature of their implementation. Building these tools is out of scope for small, start-up studios, and therefore reaching initial scale likely requires a partner that can provide not only those tools but the expertise needed to use them and interpret their output.
This dynamic obviously benefits larger organizations with the resources to staff centralized publishing teams. ATT has engendered a substantial barrier to entry for the mobile free-to-play gaming market: specialized skills that are totally unrelated to building games are needed to achieve provable traction and minimum scale, and the cost of that expertise is likely prohibitive for game studios with small portfolios, let alone single-title start-ups. And building games into billion-dollar businesses is much more difficult (if not impossible) to achieve now through direct-response advertising channels alone: cross-media marketing provides a more realistic pathway to engaging a large audience, which necessitates a larger team with more diverse capabilities in addition to more sophisticated advertising measurement solutions.
Importance of personalization for in-game content and diversity of player journeys
As something of a corrolary to the above: because Self-Attributing Networks like Meta and Google were so adept at building a feedback loop between in-game activities and advertising campaigns, mobile game developers were able to assume that the traffic they received from those channels was optimally targeted to their games on the basis of player behavioral history. For that reason, mobile free-to-play studios merely had to focus content development on a singular game experience: “the game” that the stream of curated, filtered cohorts of acquired users would play.
This changes in the post-ATT environment because the users being acquired into games can’t be assumed to be curated or filtered on the basis of past in-game behaviors. Mobile game developers can’t make assumptions about the users that are being sourced to them by Self-Attributing Networks in the same way that they previously could: players must be vetted for interest and enthusiasm within the game itself, ideally through early in-game behaviors.
As with the tools required for advertising measurement, the need for in-game personalization places a new burden on game developers. But it also unlocks an immense amount of upside opportunity: the previous game development model saw developers building singular experiences that appealed to potential audiences in a binary way, and mobile game developers effectively outsourced all personalization efforts to their advertising partners.
For the reasons outlined above, mobile game developers can’t do that any longer. Internalizing this personalization exercise increases the operational complexity of developing a game, but it also obviates the binary relevance requirement that previously dictated how games were distributed via ad channels. If game developers aren’t forced to build singular experiences that adhere strictly to the discrete audience segments that ad channels can target, then they are free to build broad, dynamic experiences that can service many segments in parallel through diverse player journeys supported by personalized content.
This, of course, expands the scope of content that must be developed for any game, even if superficially through aesthetics (art production) or simple elements of game design like level difficulty. But new tools built on generative AI can assist with content production: the real challenge with personalization is building the infrastructure needed to optimize game experiences based on player feedback and behaviors. Again: developing this infrastructure is costly and requires expertise that is not directly related or even immediately adjacent to game design (see a case study here). While I believe that personalization can more easily be achieved through generalized SaaS platforms and off-the-shelf tools than can cross-media advertising measurement, implementing these tools successfully does require in-house data science expertise.
But personalization doesn’t merely represent an added cost on building and maintaining games: it also presents upside in terms of revenue opportunity. As I detail in Freemium Economics, dynamic content experiences form the foundation of the freemium model, which, at its best, eliminates consumer surplus by availing each individual user with an optimally gratifying product experience. Personalization is the framework that provides for this. Prior to ATT, personalization was primarily the remit of ad channels, which instituted it in a constrained manner. In the post-ATT era, mobile game developers have the latitude (and burden) of implementing it themselves, to a much greater potential revenue benefit.
Focus on multi-LTV player economics
As the unit economics related to user acquisition become more challenging, achieving workable CAC/LTV math requires reducing acquisition costs, improving LTV, or both. Building products for larger and broader audiences can potentially achieve the former but can also result in impaired monetization; personalization can achieve both but requires specialized technologies and expertise, and it doesn’t necessarily close the gap. One means of improving unit economics is to deliver more than one product LTV with the acquisition of any user: to manage a user’s journey through multiple products with systematic cross-promotion across a portfolio of games.
Many developers employed a monetization calculus prior to ATT that is fundamentally similar to multi-LTV economics through persistent, “always-on” re-engagement and re-targeting. Re-engagement allows game developers to avoid the sunk acquisition cost of churn by using the data generated from user acquisition (the device identifier) to profitably extend user LTV through targeted advertising to known players. When this works (ie. is economical), it can create a recurring revenue cycle from known cohorts that surfaces new opportunities to achieve profitable CAC/LTV opportunities from an existing pool of users.
Re-engagement (also referred to as re-marketing) is difficult on iOS in the post-ATT era given limited access to the IDFA. Other user identifiers, such as email addresses, can be used to communicate with churned players to achieve re-engagement, but collecting emails in mobile games can be difficult, email addresses can be imprecise communication touchpoints, and the conversion funnel from email to app download is challenging.
Cross-promotion allows a game developer to acquire a user once and potentially establish a relationship with them across multiple titles, but it is predicated on a number of conditions. The first and most obvious (and difficult to achieve) is a game portfolio that contains enough homogenous content to support shifting users from one title to another. This is no small feat: back in 2015, I wrote about Ketchapp’s ability to do this, but Ketchapp was a hypercasual game publisher in the nascent era of in-game ads monetization. In the 2-3 years leading up to ATT, very few developers could justify moving a user from one of their apps to another given a lack of ability to accurately predict churn, insufficient portfolio size, and the skewed economics of “selling” a user to a high-ARPU game relative to the expected value of shifting that user to another game in their own portfolio.
This changes in the ATT era. I’ve written extensively about the potential for cross-promotion in years past, and the practice has only minimally been adopted within mobile gaming, but this time really is different. Cross-promotion sits at the core of the Content Fortress concept, which isn’t specific to gaming: pairing advertising infrastructure with a broad content portfolio allows companies to shift users across contexts using first-party data.
Cross-promotion clearly requires some minimal level of scale to be feasible, but given the consolidation that has already taken place within mobile gaming, there certainly exist a number of studios that could capably deploy it. My belief is that cross-promotion not only becomes a motivating factor for M&A by larger, diversified studios, but, relatedly, it inspires new studio formation as an exit path.