Moving Players Between Your Own Games: A Practical Guide to Mobile Game Cross-Promotion in 2026
As of September 2026, paid UA is getting more crowded every year. According to AppsFlyer, global gaming app UA spend reached $25 billion in 2025, up 3.8% year over year, while ad impressions rose 20% and the share of paid installs grew 10%. With more ads competing for the same inventory, the first place studios running two or more games tend to look is the inventory they already own: the traffic inside their own titles.
Cross-promotion means showing your players one of your other games. Because no media budget goes out the door, it looks free. In practice, it is paid for with ad inventory you could have sold and with part of the revenue of the game the player came from. Polish publisher BoomBit has said cross-promotion generates 13% of its profits, so done well it is a meaningful channel. Done poorly, it does not bring in new players at all; it just moves the same players around between your games.
This post extends the channel-diversification argument from Why Over-Relying on Google and Meta Is a UA Risk — And How to Build a Balanced Channel Mix and uses public data to ask how much you can actually rely on your own traffic as a channel.
How Does Mobile Game Cross-Promotion Work?
Cross-promotion runs through four main mechanisms: house ads, fixed in-game placements, cross-game rewards, and shared accounts or loyalty programs. The further down that list you go, the more it costs to build, and the more naturally it fits into the player experience.
House ads replace third-party ads in existing interstitial or banner slots with ads for your own games. Google AdMob's help documentation describes house ad campaigns as cross-promotion campaigns that serve ads in an app using the developer's own inventory for free. They are the fastest to launch, but to the player they look like any other ad.
Fixed in-game placements are spots designed as part of the game's UI, such as a "more games" menu on the main screen or an event banner. BoomBit shared that an optional games menu, which players open on their own, drives 85% of its cross-promotion installs, and stressed that cross-promotion should run as a standalone ad unit rather than inside the mediation stack. Once it is mixed into the ad auction, it becomes hard to control which impressions went to your own games, and hard to measure them.
Cross-game rewards give players a reward in game A for reaching a goal in game B. The largest-scale example is Supercell, which ran pre-registration milestone rewards across all of its games ahead of the Squad Busters launch. Check store policy before designing this. Apple's App Store Review Guideline 3.2.2(x) prohibits forcing users to download other apps in order to access functionality or content, while allowing apps to incentivize in-app actions such as completing a level. Tying the reward to play behavior in the new game rather than to the install itself is the safer choice for both policy and player quality.
Shared accounts and loyalty programs tie multiple games together under one account and one points system. In December 2024 Supercell announced Supercell ID Rewards, which lets players earn points through activity across its games and services and redeem them for exclusive skins and items in several titles. It costs more to build than individual ads, but because it turns the whole portfolio into one experience, it is the most long-term of the four.
Mechanism | How it works | Strengths | Watch out for |
|---|---|---|---|
House ads | Your own game ads shown in existing ad slots | Fast to launch, little extra development | Gives up third-party ad revenue; players see it as just another ad |
Fixed in-game placements | "More games" menus, event banners built into the UI | Players browse on their own; 85% of BoomBit's cross-promo installs | Must be separated from mediation to control and measure |
Cross-game rewards | Reward in game A for reaching a goal in game B | Strong motivation; good for launch volume | Check store policy; reward-driven installs may retain poorly |
Shared accounts and loyalty | One account and points system across games | Long-term lift to portfolio retention and conversion | High build cost; limited effect with few titles |
What Share of New Installs Can Cross-Promotion Deliver?
There is no public industry benchmark for cross-promotion's share of total installs yet, and the reliable public figures are limited to a handful of company disclosures and one academic study.
Claims that circulate online, such as "a portfolio of five or more titles generates more than 15% of new installs through cross-promotion," turn out on inspection to be framed as questions to test with data, not measured results. Rather than treating numbers like that as targets, it is safer to calibrate from data with clear sources and context, like the following.
Source | Public figures | What it tells you |
|---|---|---|
BoomBit (Gamigion podcast, Dec 2024) | Cross-promotion generates 13% of company profits; optional games menu drives 85% of cross-promo installs | At a mature publisher, profit contribution can reach double digits. Works best before scaling and in a game's sunset phase |
Gameloft study of eight games (Journal of Marketing Analytics, 2021) | Of more than 50 million ads viewed, about 139,000 led to an install of another Gameloft game | Roughly 0.28% view-to-install (calculated from the two figures). Meaningful volume requires very large impression counts |
Supercell's Squad Busters (Deconstructor of Fun, Jul 2024) | 40 million pre-registrations; about 40 million cumulative downloads in the first month (Sensor Tower estimate); more than 60% of downloads from Tier 3 countries | Supercell's biggest launch for the first five days, then fell off quickly |
What these figures share is that cross-promotion volume is capped by the source game's traffic. Daily active users in the source game, times impressions per user, times the view-to-install rate, is the ceiling, and spending more money does not raise it. An AppAgent guide based on two years of running cross-promotion at EA sets a baseline of at least 10,000 DAU for a source game to be worth using. Below that, no amount of extra impressions will produce enough volume to lift a new title.
What Does Cross-Promotion Really Cost Compared With Paid UA?
The cost of cross-promotion is not media spend but the ad revenue the source game gives up plus the in-app purchase revenue it loses. Add the two together, and depending on region and placement, cross-promotion can cost as much as paid CPI, or more.
The first cost is opportunity cost. Every interstitial slot used for a house ad is a slot that could have earned third-party ad revenue. Converted to a per-install figure, that cost equals the placement's eCPM (ad revenue per 1,000 impressions) divided by the view-to-install rate. The table below applies the roughly 0.28% view-to-install rate from the Gameloft study to the regional interstitial eCPM ranges AdReact published in May 2026. Conversion rates vary widely by game and placement, so read these as a way to compare regions rather than as absolute values.
Source game placement (region) | Interstitial eCPM (AdReact, 2026) | Ad revenue forgone per cross-promo install (calculated) |
|---|---|---|
US and Canada, video | $12-25 | about $4.30-9.00 |
US and Canada, static or playable | $6-12 | about $2.20-4.30 |
Western Europe, video | $8-18 | about $2.90-6.50 |
Western Europe, static | $4-10 | about $1.40-3.60 |
Southeast Asia, video | $2-6 | about $0.70-2.20 |
Southeast Asia, static | $1-3 | about $0.40-1.10 |
Put those next to current paid UA prices and the decision rule becomes visible. Admiral Media's April 2026 CPI ranges for Tier 1 gaming markets look like this.
Genre (Tier 1 markets) | Android CPI | iOS CPI |
|---|---|---|
Hypercasual | $0.30-1.20 | $0.80-2.50 |
Casual puzzle | $1.20-3.00 | $2.50-6.00 |
Idle RPG | $2.00-5.00 | $4.00-9.00 |
Mid-core strategy | $4.00-10.00 | $8.00-18.00 |
If you cross-promote a casual puzzle game through US video interstitials, the ad revenue forgone per install (about $4.30-9.00) can exceed the paid Android CPI for the same genre ($1.20-3.00). Promote a mid-core strategy game through low-eCPM Southeast Asian placements instead, and cross-promotion comes out far cheaper than paid UA. The claim that cross-promotion is free is mostly true only when you promote a high-CPI genre from low-value ad inventory.
The second cost is lost revenue in the source game, covered in the next section. On top of that, you also need to compare how long cross-promoted players stay relative to paid players before you know the real cost per retained player. Comparing install prices alone creates the illusion that cross-promotion is always cheap.
Why Doesn't Cross-Promotion Grow Your Player Base? The Limits of Cannibalization and Audience Overlap
Cross-promotion is a tool for redistributing players you already have across your portfolio, and it comes with cannibalization: the source game loses revenue as players move to the new one.
The most rigorous public measurement of this effect is a 2021 paper that analyzed data from eight Gameloft games. The researchers used propensity score matching on more than 50 million ad views, and the roughly 139,000 installs they produced, to correct the observational data so it mimics a randomized experiment. Overall, players who installed the advertised game showed a 20.66% decrease in expected IAP revenue in the game they were already playing. The drop was larger in higher-revenue games and did not appear to depend on how much the player had spent before seeing the ad. In Gameloft's case, revenue gained in the newly installed games exceeded the loss in the original games, producing a 17.69% increase in expected revenue across the portfolio. The authors conclude that companies need to choose carefully which game they promote to fully benefit from cross-promotion.
Audience overlap shows up even more clearly at launch. Supercell ran pre-registration rewards across all of its games before Squad Busters launched, and pre-registrations reached 40 million. First-month cumulative downloads, by Sensor Tower's estimate, were also about 40 million. But as Deconstructor of Fun's analysis points out, the approach attracted an audience that was already happy in a Supercell game, and more than 60% of downloads came from Tier 3 countries, mirroring the player distribution of Supercell's existing titles. Other Supercell games saw little to no DAU drop during the launch, which suggests many players tried the new game without leaving their current one. For its first five days Squad Busters was Supercell's biggest launch, but downloads fell off quickly after that, and although the game earned more than $100 million in its first seven months, Supercell announced the end of its development on October 30, 2025.
Taken together, the two cases show where cross-promotion fits and where it does not. Moving players who already show churn signals, non-payers, or players of a game in its sunset phase into a new title does a lot to keep them inside your portfolio. This connects to the lifecycle design discussed in Lifecycle Marketing in Mobile Games: Why the Lever 74% of Studios Trust Feels Stale to 80% of Them, and the timing of catching players on their way out follows the same principles as Winning Back Lapsed Mobile Game Players: A Re-engagement Framework That Works. Repeatedly showing a new game to the core payers of a high-revenue title, on the other hand, is the most expensive cross-promotion there is. The AppAgent guide recommends capping spenders at three impressions per week and non-spenders at ten for exactly this reason: to avoid moving high-revenue players into games that monetize less.
The most important limit is that cross-promotion alone cannot tell you whether a new game works for players outside your portfolio. Your own players already know your brand, art style, and controls, so their early metrics tend to look better than those of outside players. In a new game's soft launch, always split cross-promoted traffic from external traffic when reading the numbers.
How Should You Measure the Incremental Impact of Cross-Promotion?
Judge cross-promotion not by installs in the destination game but by the incremental difference in portfolio-wide revenue and retention between players who saw cross-promotion and players who did not.
Attribution tools credit every install that came through a cross-promotion placement to cross-promotion. But your own players are exactly the group most likely to hear about your new game without an ad, so a sizable share of those installs may have happened anyway. Meanwhile, the revenue lost in the source game never appears in an attribution report. Install-based reporting inflates the gain and erases the cost.
The most reliable method is a randomized holdout. Randomly select a portion of the source game's players, exclude them from every cross-promotion placement, and compare them with everyone else. The core metrics are portfolio-wide revenue per player (source game plus destination game), total active days across the portfolio, and portfolio churn, with source-game revenue tracked separately to size the cannibalization. Look at longer windows such as D30 and D90 rather than the first few days after install; the net effect of cannibalization and migration only shows up over time. As BoomBit emphasized, running cross-promotion as a standalone placement separate from mediation is what makes this kind of group split possible.
If a randomized holdout is not feasible, the next best option is what the Gameloft study did: use propensity score matching to pair players who installed with similar players who did not. How to set the holdout share, sample size, and test length is covered in How to Design an Incrementality Test for Mobile Game UA: Holdouts, Geo-Lift, Sample Size, and Reading the Result, and the same design applies directly to cross-promotion.
Results are far more useful when read by segment, because incrementality differs between payers and non-payers, between source games in growth and in decline, and by how far the destination game's genre is from the source game's. Rules built on those results, deciding which players see which game, are what turn cross-promotion from a vaguely free channel into a UA channel you can actually manage.
What Can a Single-Title Studio Do Instead?
A studio with only one game can borrow the effect of cross-promotion by partnering with other studios' traffic, with communities, or with platforms, instead of relying on a portfolio of its own.
The most direct route is a cross-promotion exchange with another studio. Each side shows the other's game in its own in-game placements, trading traffic instead of money. Overlapping genre tastes raise conversion (the AppAgent guide likewise recommends promoting puzzle games in puzzle games and strategy in strategy within a portfolio), but if your partner is a direct competitor fighting for the same players' time and spend, the cannibalization described above hits both studios. The ideal partner shares player tastes but differs enough in session timing or monetization model that neither side's revenue takes a large hit. Agree on the exchange ratio in terms of installs or D7-retained players rather than impressions, and make sure both sides can verify results through their own mobile measurement partner (MMP) tracking links, so there is nothing to dispute later.
The second route is community and IP collaboration: running collab events with games whose players have similar tastes, or linking community channels such as Discord servers to co-host events. How studios use community channels before and after launch is covered in Discord Marketing for Game Studios: What a Server Does From Pre-Launch Hype to LiveOps Retention.
The third route is partnering with platforms where gamers discover multiple games. It borrows, through an external gamer community, the structure large publishers enjoy inside their own portfolios: recommending the next game to players who already enjoy playing games. What matters here is whether those players actually play, not how many install, so pricing tied to play behavior rather than installs is the better fit with the quality principles of cross-promotion.
Recommending the Next Game to Committed Gamers Beyond Your Portfolio: Playio's Perspective
Cross-promotion's biggest limit is that it only works within the closed pool of your own players, and Playio's role is to widen that pool to a community of five million gamers.
Playio is a community space where gamers talk about games day to day and discover new ones. Because AI analyzes genre preferences, play history, and in-game behavior data to match gamers with games that fit their tastes, even a studio without a portfolio can do what large publishers do inside theirs: recommend the next game to players who loved a similar one. For single-title studios it is an alternative to a cross-promotion exchange, and for multi-title studios it adds a source of genuinely new players that redistributing your own players cannot provide.
CPI is the default pricing model, and CPE tied to reaching a playtime threshold or completing a specific in-game action is also supported. Missions aimed at players who have not yet installed the game can run alongside missions that deepen engagement among players who already play it, so new-player acquisition and existing-player activation can be designed within a single campaign.
You can find more details here. (https://playioadsen.oopy.io/bizdeck)
Key Takeaways
As of September 2026, rising competition in paid UA has made cross-promotion through your own game traffic more valuable, but the idea that it is a free channel is only half true. Of the four mechanisms, house ads, fixed in-game placements, cross-game rewards, and shared accounts or loyalty programs, placements that players browse on their own drive the most installs, and BoomBit now earns 13% of its profits from cross-promotion this way. There is no public benchmark for cross-promotion's share of installs, and volume can never exceed the source game's traffic. Converted into forgone ad revenue, a cross-promotion install in high-eCPM regions can cost more than paid CPI, and cannibalization must be counted too: in the Gameloft study, expected IAP revenue in the source game fell by about 20%. Because cross-promotion redistributes existing players rather than adding new ones, focus it on players about to churn and on non-payers, and measure portfolio-wide incrementality with randomized holdouts. Studios with a single game can borrow the same effect through exchanges with non-competing studios, community and IP collaborations, and partnerships with gamer platforms.
For inquiries about Playio's advertising solutions, reach out at: [email protected]
Sources
AppsFlyer, State of Gaming for Marketers 2026 (January 14, 2026; $25B gaming UA spend in 2025, +3.8% YoY; paid install share +10%; ad impressions +20%): https://www.appsflyer.com/company/newsroom/pr/gaming-marketing/
Gamigion, Cross-Promotion for Mobile Games That Works (Felix Braberg, December 17, 2024; cross-promotion generates 13% of BoomBit's profits; optional games menu drives 85% of cross-promo installs; most effective pre-scaling and in the sunset phase; standalone ad unit rather than mediation): https://www.gamigion.com/cross-promotion-for-mobile-games-that-works/
Débordès, Caporossi, Larocque, Is my cross-promotion profitable? Evaluation of game-to-game cannibalization in free-to-play mobile games, Journal of Marketing Analytics 9(3), 2021 (eight Gameloft games; 50M+ ads viewed, 139K+ installs; propensity score matching; 20.66% decrease in expected IAP revenue in the source game; 17.69% increase in expected portfolio revenue; larger decrease in higher-revenue games): https://link.springer.com/article/10.1057/s41270-021-00122-x (abstract: https://ideas.repec.org/a/pal/jmarka/v9y2021i3d10.1057_s41270-021-00122-x.html)
Deconstructor of Fun, Squad Busters One Month Later: Down, But Not Out (July 8, 2024; 40M pre-registrations; about 40M first-month cumulative downloads per Sensor Tower estimate; Supercell's biggest launch for the first five days; 60%+ of downloads from Tier 3; pre-registration rewards in every Supercell game; little to no DAU drop in other Supercell titles; major existing titles at 50%+ Tier 3 DAU share): https://www.deconstructoroffun.com/blog/2024/7/8/squad-busters-one-month-later-down-like-a-lot-but-not-out
Game World Observer, "We Were Wrong": Supercell to Shut Down Squad Busters (October 30, 2025; $100M+ revenue in first seven months; 60M downloads by May 2025; final update December 2025): https://gameworldobserver.com/2025/10/30/we-were-wrong-supercell-to-shut-down-squad-busters-two-years-after-games-release
Supercell, Leveling Up the Player Experience with Supercell ID Rewards (December 12, 2024): https://supercell.com/en/news/id-rewards/
Google AdMob Help, About AdMob campaigns (house ad campaigns as free cross-promotion using the developer's own inventory): https://support.google.com/admob/answer/6162747?hl=en
Apple, App Store Review Guidelines 3.2.2(x) (apps must not force users to download other apps; incentivizing in-app actions such as completing a level is allowed): https://developer.apple.com/app-store/review/guidelines/
AppAgent, A Guide to Cross-Promotion: How to Increase Installs for Your Mobile Game for Free (Clara Popescu, published April 22, 2020, updated May 14, 2025; two years of cross-promotion at EA; 10K+ DAU source game; example caps of three weekly impressions for spenders and ten for non-spenders; same-genre promotion recommended): https://appagent.com/blog/a-guide-to-cross-promotion-how-to-increase-installs-for-your-mobile-game-for-free/
AdReact, Interstitial Ad Best Practices for Mobile Games (May 8, 2026; regional interstitial eCPM ranges): https://adreact.com/blog/interstitial-ad-best-practices-mobile-games/
Admiral Media, Mobile App Marketing Benchmarks 2026 (April 19, 2026; Tier 1 Android and iOS CPI ranges by genre): https://admiral.media/mobile-app-marketing-benchmarks-2026/
Ad revenue forgone per install: calculated as eCPM ÷ (1,000 × 0.278%), where 0.278% is about 139,000 installs ÷ about 50 million ad views from the Gameloft study (both reported as lower bounds, so the rate is approximate)