Gaming Report

Gaming’s Personalization Imperative

The next growth lever isn’t more players. It’s understanding and engaging your players better than anyone else.

著者：Yohann Plantec and Anders Christofferson

First published on 8月 18, 2026
min read


概要 Amid slower industry growth, boosting player engagement through personalization is becoming more important. Most gamers are comfortable with or neutral toward personalized offers based on their in-game activity, a Bain survey found. The required AI technology for personalization has improved: One early adopter lifted a LiveOps campaign’s revenue per player by more than half. Winning depends on having clear goals, cross-functional ownership, the right tech partners, and the courage to start before the solution is perfect. This article is part of Bain's 2026 Gaming Report. Explore the report The gaming industry has weathered a tumultuous past few years. The pandemic boom gave way to a consolidation wave. Now, the industry is settling into its new normal: It’s harder than ever to win new players and keep engaging existing ones. While some pockets of growth remain, the across-the-board expansion of the past decade has stalled. Global gaming software revenue has grown at a compound annual rate of about 3% over the past four years, and Bain forecasts a similar pace over the next four years (see Figure 1). Meanwhile, gaming’s share of total media consumption has plateaued. Most of the audience has been captured; the hours are largely spoken for. Figure 1 Global gaming software revenue is expected to continue growing by about 3% annually, in line with recent years Notes: Forecast as of December 2025; excludes hardware, advertising, and esports revenue Sources: IDC; PwC; Omdia; S&P Capital IQ; Bain analysis This isn’t a secret. The past few years brought layoffs, studio closures, dried-up funding for new games, and delayed major releases. Consolidation has accelerated. AI is arriving, promising transformation but also raising uncomfortable questions and not yet delivering full value. The postpandemic growth that executives planned for five years ago hasn’t panned out. Even worse, many companies haven’t adjusted their strategies accordingly, which will lead to weaker results. Most executives sense that something structural has shifted. What fewer have internalized is the logical conclusion: If the overall market isn’t growing as fast as before, then strengthening relationships with the players you already have becomes even more crucial to long-term success. Personalization—customizing communications, advertisements, and content to each player—is one of the most important ways that gaming companies can boost engagement. Personalization pays off on two fronts at once: Players get a more enjoyable, tailored experience, and that improved experience can deliver the stronger engagement and revenue that studios are ultimately pursuing. Gaming companies understandably pour significant time, energy, and resources into perfecting gameplay. Now, it’s time for them to invest more in making everything outside of gameplay an amazing experience, too. That’s where technology-enabled personalization comes in; it can help companies win players’ hearts and minds. The value is waiting to be captured Many executives may be surprised to learn that players are actually open to data-driven personalization. Bain’s latest annual survey found that most gamers feel either comfortable with or neutral toward personalized offers based on their in-game activity (see Figure 2). Figure 2 The vast majority of gamers feel comfortable with or neutral toward personalized offers across different scenarios Notes: Hypothetical scenario A was “After noticing you’d been browsing skins for a particular character, the game offers you a 20% discount on one”; scenario B was “Based on the games you play across multiple platforms and devices, the game recommends an item suited to your overall play style”; scenario C was “The game offers you a discount on a key upgrade exactly at the moment its system predicts you might stop playing” Source: Bain Media Consumption Survey, June 2026 (n=5,339) True one-to-one personalization has been a holy grail for decades. Other consumer categories have shown what the playbook could look like in gaming. Video streaming platforms use viewing history, pause behavior, and browsing patterns to recommend a customer’s next binge-worthy title. E-commerce marketplaces serve up more relevant goods based on each person’s past searches and purchases. Gaming companies that have direct relationships with their players at scale are sitting on rich behavioral data. For example, companies that run their own gaming platform or operate a direct-to-consumer subscription service have the raw material in the form of in-game micro-decisions, social graph connections, and purchase history. Until recently, the challenge was that the technology to deliver one-to-one personalization for gamers didn’t exist. Now, that’s starting to change. What one-to-one personalization actually means, and why now The term “personalization” gets used loosely, so it’s worth being precise. Studios have run customer segmentation plays for years: group players into cohorts, send targeted messages, run A/B tests to compare different offers. That’s not what we’re talking about. True one-to-one personalization means every interaction is optimized in real time for the individual player, based on their specific behavior, preferences, and stage of the customer life cycle—without a human analyst making every decision. This is made possible by contextual bandit algorithms, a class of machine learning models that enable agents to learn continuously from player behavior and optimize decisions in real time. The agent treats every player according to what the data says will work best for that individual. It gears a high-frequency player’s experiences toward retaining them and deepening their engagement. For example, a developer advertising a game’s new season could send a personalized email to a casual player highlighting the return of that player’s favorite character. Meanwhile, a hardcore player might receive an in-game notification about the competitive ranking system launching with the new season. Contextual bandit engines are still frontier technology, but the barrier to entry is getting lower. The tooling has improved, the models have gotten better at inferring patterns from incomplete data, and technology partners that can handle much of the orchestration layer have emerged. “The technology has completely changed over the last two years and is now able to deliver real value at a reasonable cost,” says George Khachatryan, vice president and head of AI decisioning at Braze, one of the technology partners that Bain works with to help gaming companies implement these kinds of solutions. Nevertheless, only a small number of gaming companies are beginning to build the infrastructure. Most industry participants are in wait-and-see mode. Contextual bandit engines are still frontier technology, but the barrier to entry is getting lower. Five places to put one-to-one personalization to work Outside gameplay itself, the highest-impact personalization opportunities cluster around five areas. We’ve roughly sequenced them by where ROI tends to show up first. In-game store and offer optimization. This includes dynamic pricing and tailored bundles based on each player’s spending history and propensity to spend. It’s typically the fastest path to measurable revenue growth, and it’s the area where studios with good transaction data can move quickly. Lapsed player reengagement. Reengagement is one of the highest-ROI use cases because the players already know the game. This category includes intelligent campaigns that identify when a player is drifting or has stopped playing a game, determine the right moment and channel to reach them (email, push, in-game), and deliver a message or offer tuned to what attracted them before. Personalized onboarding. Most studios still deliver the same onboarding to every player. The first few sessions playing a game are disproportionately predictive of long-term retention. By analyzing each person’s gameplay during those initial sessions, companies can build adaptive onboarding mechanisms outside the game—personalized hints and tips, new feature recommendations, and so on—that reflect player skill level, play style, and preferences. In-game content discovery. The goal is to surface the right modes, maps, and events for each player at the right moment. This can reduce the friction between logging in and finding something worth doing. Cross-title recommendations within a portfolio. For publishers with multiple titles, using behavioral data from one game to intelligently guide players toward others is a big opportunity. This is where scale creates a compounding advantage: The more titles you have, the more signal you have, and the better your recommendations become. The real power of contextual bandit technology is that agents could assess trade-offs between these five different areas in real time—such as deciding that the best way to engage and monetize a player is to offer that person a bundle vs. sending an in-game recommendation or suggesting a new, similar game in the store. For example, by implementing Braze’s solution, a large free-to-play gaming company replaced its fixed-schedule, broadly targeted promotions with real-time offers personalized to each player. Powered by AI decision making built on reinforcement-learning (contextual bandit) engines, the system experimented across thousands of unique trigger-and-offer combinations. It automatically matched a player’s in-game behavior—for example, reaching a gameplay milestone—to the offer deemed most likely to convert them from a free to a paying player. Dozens of customer features were fed into the system to inform each decision. The approach boosted the in-game LiveOps campaign’s revenue per player by more than half. What it takes to get this right The organizational requirements are as important as the technical ones. The most successful companies will follow an intentional, holistic process—setting goals and measurable metrics, deeply researching the technology’s capabilities, making build vs. buy decisions, running pilots, and executing the plan to scale the capabilities across the organization. In our work with gaming companies around the world, we’ve seen a few common pitfalls. Here are the four most important things to get right. Clearly define the desired outcomes. Without clear goals, it’s impossible to develop an effective plan and measure progress against it. For example, is the company trying to improve player satisfaction scores, increase the number of hours played, optimize discovery of new games and in-game content, or increase the amount of revenue per player generated over time? Setting clear objectives is the only way to maximize this technology’s potential. Tap the right people. The most successful companies will treat personalization as a cross-functional capability, not a marketing tool. It touches product, monetization, community, and player support. The studios that get this right will have a dedicated function with executive sponsorship, not a team tucked under the CMO. Who are the enthusiastic people best equipped to champion this initiative and make sure it succeeds? Find the right partners. This is bleeding-edge technology, and most gaming companies likely can’t do it all themselves. In particular, the engagement orchestration layer—the capability that sequences and delivers personalized interactions across surfaces and channels—is an area where partnering typically makes more sense than building from scratch. All five use cases outlined above depend on managing the timing and sequencing of touchpoints so that the player experience feels coherent rather than noisy. Get started now. At this challenging moment for the gaming industry, executives might feel tempted to wait for someone else to prove the model before jumping in. They might also feel nervous about applying AI to the customer experience, and so they’re waiting for the technology to fully mature. That’s a mistake. Early movers aren’t just gaining a capability; they’re building a data flywheel. These early adopters recognize that the technology is good enough now and improving extremely quickly, and they plan to evolve with it. Those that wait for perfection will get left behind. Read the Next Chapter Game Distribution: Half Your Players Already Buy Direct Read our 2026 Gaming Report Download the PDF Explore the report Explore more Gamer Survey: The End of the Average Gamer AI Won’t Save an Unfocused Game Gaming’s Personalization Imperative Game Distribution: Half Your Players Already Buy Direct

概要

Amid slower industry growth, boosting player engagement through personalization is becoming more important.
Most gamers are comfortable with or neutral toward personalized offers based on their in-game activity, a Bain survey found.
The required AI technology for personalization has improved: One early adopter lifted a LiveOps campaign’s revenue per player by more than half.
Winning depends on having clear goals, cross-functional ownership, the right tech partners, and the courage to start before the solution is perfect.

The gaming industry has weathered a tumultuous past few years. The pandemic boom gave way to a consolidation wave. Now, the industry is settling into its new normal: It’s harder than ever to win new players and keep engaging existing ones.

While some pockets of growth remain, the across-the-board expansion of the past decade has stalled. Global gaming software revenue has grown at a compound annual rate of about 3% over the past four years, and Bain forecasts a similar pace over the next four years (see Figure 1). Meanwhile, gaming’s share of total media consumption has plateaued. Most of the audience has been captured; the hours are largely spoken for.

visualization — Global gaming software revenue is expected to continue growing by about 3% annually, in line with recent years

This isn’t a secret. The past few years brought layoffs, studio closures, dried-up funding for new games, and delayed major releases. Consolidation has accelerated. AI is arriving, promising transformation but also raising uncomfortable questions and not yet delivering full value. The postpandemic growth that executives planned for five years ago hasn’t panned out. Even worse, many companies haven’t adjusted their strategies accordingly, which will lead to weaker results.

Most executives sense that something structural has shifted. What fewer have internalized is the logical conclusion: If the overall market isn’t growing as fast as before, then strengthening relationships with the players you already have becomes even more crucial to long-term success.

Personalization—customizing communications, advertisements, and content to each player—is one of the most important ways that gaming companies can boost engagement. Personalization pays off on two fronts at once: Players get a more enjoyable, tailored experience, and that improved experience can deliver the stronger engagement and revenue that studios are ultimately pursuing.

Gaming companies understandably pour significant time, energy, and resources into perfecting gameplay. Now, it’s time for them to invest more in making everything outside of gameplay an amazing experience, too. That’s where technology-enabled personalization comes in; it can help companies win players’ hearts and minds.

The value is waiting to be captured

Many executives may be surprised to learn that players are actually open to data-driven personalization. Bain’s latest annual survey found that most gamers feel either comfortable with or neutral toward personalized offers based on their in-game activity (see Figure 2).

True one-to-one personalization has been a holy grail for decades. Other consumer categories have shown what the playbook could look like in gaming. Video streaming platforms use viewing history, pause behavior, and browsing patterns to recommend a customer’s next binge-worthy title. E-commerce marketplaces serve up more relevant goods based on each person’s past searches and purchases.

Gaming companies that have direct relationships with their players at scale are sitting on rich behavioral data. For example, companies that run their own gaming platform or operate a direct-to-consumer subscription service have the raw material in the form of in-game micro-decisions, social graph connections, and purchase history.

Until recently, the challenge was that the technology to deliver one-to-one personalization for gamers didn’t exist. Now, that’s starting to change.

What one-to-one personalization actually means, and why now

The term “personalization” gets used loosely, so it’s worth being precise. Studios have run customer segmentation plays for years: group players into cohorts, send targeted messages, run A/B tests to compare different offers. That’s not what we’re talking about. True one-to-one personalization means every interaction is optimized in real time for the individual player, based on their specific behavior, preferences, and stage of the customer life cycle—without a human analyst making every decision.

This is made possible by contextual bandit algorithms, a class of machine learning models that enable agents to learn continuously from player behavior and optimize decisions in real time. The agent treats every player according to what the data says will work best for that individual. It gears a high-frequency player’s experiences toward retaining them and deepening their engagement. For example, a developer advertising a game’s new season could send a personalized email to a casual player highlighting the return of that player’s favorite character. Meanwhile, a hardcore player might receive an in-game notification about the competitive ranking system launching with the new season.

Contextual bandit engines are still frontier technology, but the barrier to entry is getting lower. The tooling has improved, the models have gotten better at inferring patterns from incomplete data, and technology partners that can handle much of the orchestration layer have emerged.

“The technology has completely changed over the last two years and is now able to deliver real value at a reasonable cost,” says George Khachatryan, vice president and head of AI decisioning at Braze, one of the technology partners that Bain works with to help gaming companies implement these kinds of solutions.

Nevertheless, only a small number of gaming companies are beginning to build the infrastructure. Most industry participants are in wait-and-see mode.

Contextual bandit engines are still frontier technology, but the barrier to entry is getting lower.

Five places to put one-to-one personalization to work

Outside gameplay itself, the highest-impact personalization opportunities cluster around five areas. We’ve roughly sequenced them by where ROI tends to show up first.

In-game store and offer optimization. This includes dynamic pricing and tailored bundles based on each player’s spending history and propensity to spend. It’s typically the fastest path to measurable revenue growth, and it’s the area where studios with good transaction data can move quickly.
Lapsed player reengagement. Reengagement is one of the highest-ROI use cases because the players already know the game. This category includes intelligent campaigns that identify when a player is drifting or has stopped playing a game, determine the right moment and channel to reach them (email, push, in-game), and deliver a message or offer tuned to what attracted them before.
Personalized onboarding. Most studios still deliver the same onboarding to every player. The first few sessions playing a game are disproportionately predictive of long-term retention. By analyzing each person’s gameplay during those initial sessions, companies can build adaptive onboarding mechanisms outside the game—personalized hints and tips, new feature recommendations, and so on—that reflect player skill level, play style, and preferences.
In-game content discovery. The goal is to surface the right modes, maps, and events for each player at the right moment. This can reduce the friction between logging in and finding something worth doing.
Cross-title recommendations within a portfolio. For publishers with multiple titles, using behavioral data from one game to intelligently guide players toward others is a big opportunity. This is where scale creates a compounding advantage: The more titles you have, the more signal you have, and the better your recommendations become.

The real power of contextual bandit technology is that agents could assess trade-offs between these five different areas in real time—such as deciding that the best way to engage and monetize a player is to offer that person a bundle vs. sending an in-game recommendation or suggesting a new, similar game in the store.

For example, by implementing Braze’s solution, a large free-to-play gaming company replaced its fixed-schedule, broadly targeted promotions with real-time offers personalized to each player. Powered by AI decision making built on reinforcement-learning (contextual bandit) engines, the system experimented across thousands of unique trigger-and-offer combinations. It automatically matched a player’s in-game behavior—for example, reaching a gameplay milestone—to the offer deemed most likely to convert them from a free to a paying player. Dozens of customer features were fed into the system to inform each decision. The approach boosted the in-game LiveOps campaign’s revenue per player by more than half.

What it takes to get this right

The organizational requirements are as important as the technical ones. The most successful companies will follow an intentional, holistic process—setting goals and measurable metrics, deeply researching the technology’s capabilities, making build vs. buy decisions, running pilots, and executing the plan to scale the capabilities across the organization. In our work with gaming companies around the world, we’ve seen a few common pitfalls. Here are the four most important things to get right.

Clearly define the desired outcomes. Without clear goals, it’s impossible to develop an effective plan and measure progress against it. For example, is the company trying to improve player satisfaction scores, increase the number of hours played, optimize discovery of new games and in-game content, or increase the amount of revenue per player generated over time? Setting clear objectives is the only way to maximize this technology’s potential.

Tap the right people. The most successful companies will treat personalization as a cross-functional capability, not a marketing tool. It touches product, monetization, community, and player support. The studios that get this right will have a dedicated function with executive sponsorship, not a team tucked under the CMO. Who are the enthusiastic people best equipped to champion this initiative and make sure it succeeds?

Find the right partners. This is bleeding-edge technology, and most gaming companies likely can’t do it all themselves. In particular, the engagement orchestration layer—the capability that sequences and delivers personalized interactions across surfaces and channels—is an area where partnering typically makes more sense than building from scratch. All five use cases outlined above depend on managing the timing and sequencing of touchpoints so that the player experience feels coherent rather than noisy.

Get started now. At this challenging moment for the gaming industry, executives might feel tempted to wait for someone else to prove the model before jumping in. They might also feel nervous about applying AI to the customer experience, and so they’re waiting for the technology to fully mature.

That’s a mistake. Early movers aren’t just gaining a capability; they’re building a data flywheel. These early adopters recognize that the technology is good enough now and improving extremely quickly, and they plan to evolve with it. Those that wait for perfection will get left behind.

Read the Next Chapter

Game Distribution: Half Your Players Already Buy Direct

Explore more

著者

Yohann Plantec

パートナー, Los Angeles
Anders Christofferson

パートナー, Denver

Gaming’s Personalization Imperative

Gaming’s Personalization Imperative