Gamesforum Interviews Quago’s Co-founder Dan Blechner

Taking a break from the busy Gamesforum Hamburg, Dan Blechner, our Co-founder and CPO, was interviewed about Quago’s mission to empower gaming studios with deeper insights into their players. Speaking with Gamesforum’s Josh Vowles, Dan provided a glimpse into our cutting-edge sensor-data technology that offers an exclusive data layer and unique insights into player behavior.

He also shared strategies for combating various types of fraud: click fraud, in-app CPA fraud, and end-user fraud. He discussed how cheating impacts monetization margins and ways to manage cheating to ensure fair play and reliable data analytics for gaming companies. Dan emphasized that Quago’s technology not only detects post-install event abuse and cheating but also prioritizes user privacy, ensuring a secure and fair gaming environment.

Dan addressed key challenges in mobile game growth, including evolving privacy regulations and fraudulent activities that distort data and impact user acquisition. Dan explained how Quago’s advanced churn prediction model forecasts player churn within 24-48 hours, enabling timely interventions to boost retention.

Read the full interview here.

The cost of cheating in online gaming: How it affects the industry

Multiple recent research have found that 60% of online gamers feel that their multiplayer gaming experience has been negatively impacted due to other cheating players.

In the world of online gaming, cheating is like that annoying kid at the playground who keeps bending the rules to win. It’s not cool, it’s not fair, and it ruins the fun for everyone else. But in this article, we discuss how cheating leads to far-reaching consequences in the gaming industry as a whole.

The costs of cheating span from industry efforts to detect and limit it, to losses in monetization to brand reputation issues and to victims’ monetary losses due to scams. 

 Cheating is prevalent in most multiplayer online games, though measuring its full extent remains challenging. Various methods of cheating in online games can take the form of software (scripts, automation, moding) assistance or various forms of unsporting play taking advantage of exploits within the game. The Internet and darknets offer many tools and methodologies built to cheat in online games, such as this content piece.

A closer look into the online gaming cheating phenomena

The evolution of cheating in online gaming:

From classic aimbots and wall hacks to sophisticated cheat detection systems, the cat-and-mouse game between developers and cheaters has reached new heights. 

In the early days of online gaming, cheating was primarily limited to exploiting glitches and using rudimentary aimbots. These methods were relatively easy to detect and counter, and game developers quickly implemented measures to address these vulnerabilities.

As multiplayer and esports games gain prominence, the methods of cheating have become more intricate, enforcing constant innovation in anti-cheat measures. 

Cheaters began to utilize software that could analyze game data and make automated decisions, giving them superhuman reflexes and unbeatable aim. These advancements made cheating more difficult to detect and counter, as the lines between legitimate player behavior and automated cheating became increasingly blurred.

Maybe the most important aspect of this phenomenon is the fact that cheaters are constantly adapting their tactics to overcome anti-cheat technologies. 


Let’s draw a timeline of the evolution of cheating: 

  1. Early Days:
  • Methods: Glitch exploitation, basic aimbots.
  • Detection and Countermeasures: Developers quickly respond to glitch exploitation, implementing measures to fix bugs and vulnerabilities.
  1. Growth of Multiplayer games (~2010):
  • Methods: Introduction of more advanced aimbots, and wallhacks.
  • Detection and Countermeasures: Developers enhance anti-cheat measures as multiplayer gaming gains popularity. Early stages of signature-based detection.
  1. Esports Emergence (~2010+):
  • Methods: Automated decision-making software for superhuman reflexes.
  • Detection and Countermeasures: Anti-cheat measures evolve with a focus on heuristic analysis. Increased sophistication in cheat tactics requires constant innovation in detection systems.
  1. Technological Advancements (2015 and onwards):
  • Methods: Machine learning-powered automation, cloud-based cheating services.
  • Detection and Countermeasures: Developers integrate machine learning algorithms into anti-cheat systems. Cloud-based cheating services pose new challenges.
  1. Present Day (2023):
  • Methods: Device sensor-based analytics, adaptive cheats using machine learning.
  • Detection and Countermeasures: Introduction of real-time sensor-based analytics for immediate identification of suspicious behavior. Ongoing arms race with machine learning-powered cheats.
  1. Future Challenges (Beyond 2023):
  • Anticipated Methods: Continued use of machine learning-powered cheats, challenges in cloud gaming and cross-platform play.
  • Evolution in Detection: Enhanced sensor-based ML analytics to counter emerging threats. Focus on minimizing false positives through robust evidence collection.



Common types of cheating: 

Software-based cheats:

  • Aimbots: These programs automatically aim at opponents, giving players an unfair advantage in shooting games.

  • Wallhacks: These enable players to see opponents through walls.

  • Speedhacks: These programs increase the speed of a player’s character, making them difficult to catch.

  • Lag switches: These programs artificially induce lag, causing the player’s character to appear to teleport around the map, making them difficult to hit.

Exploitation of game mechanics:

  • Glitching: This involves taking advantage of unintended bugs or glitches in the game to gain an unfair advantage.

  • Boosting: This involves collaborating with other players to artificially inflate one player’s rank or statistics.

  • Farming: This involves using automated scripts or bots to perform repetitive tasks in the game, such as gathering resources or killing enemies.

Hardware-based cheats:

  • Game console mods: These mods allow players to install custom software on their game consoles, which can be used to cheat.

  • Input modifiers: These devices allow players to input commands into the game that are not normally possible, such as rapid fire or macros.

  • Memory editors: These devices allow players to modify the game’s memory in real time, which can be used to cheat.

Other, more sophisticated methods include: 

  • Data Manipulation: Cheaters can manipulate game data packets to gain an unfair advantage. This involves intercepting and modifying data exchanged between the game server and the player’s device. By altering data related to player positions, resources, or inventory, cheaters can gain control over the game world and manipulate outcomes.
  • Reverse Engineering and Code Injection: Cheaters with programming expertise may resort to reverse engineering game code to identify vulnerabilities and inject malicious code. This injected code can alter game mechanics, bypass anti-cheat measures, or grant cheaters unauthorized privileges.
  • Machine Learning-Powered automation: Advanced cheaters employ machine learning algorithms to create sophisticated software that can learn and adapt to game environments. These cheats can mimic human behavior, making them more difficult to detect by traditional anti-cheat software.
  • Cloud-Based Cheating Services: Cheaters can utilize cloud-based services to access remote cheating tools and resources. These services may provide access to aimbots, wallhacks, or other cheating software, making it easier for cheaters to evade detection.
  • DDoS (Distributed Denial-of-Service) Attacks: Cheaters may launch DDoS attacks to disrupt game servers and gain an unfair advantage. These attacks overwhelm servers with traffic, causing them to crash or become unresponsive, giving cheaters a temporary advantage in the game.
  • Social Engineering and Collusion: Cheaters may employ social engineering tactics to trick other players into revealing sensitive information or gain access to their accounts. This information can then be used to exploit vulnerabilities or manipulate game mechanics.
  • Hardware Tampering: Cheaters may tamper with hardware components, such as game controllers or input devices, to gain an unfair advantage. This may involve modifying hardware to achieve rapid-fire capabilities or altering input signals to manipulate game mechanics.


To summarize this part, even the top industry gaming professionals, such as game managers, product managers, and alike are not aware of all these methods described above or the extent to which they are used. 


Real time examples for major cheating incidents covered by the media, include: 

Call of Duty


The publisher has taken action against a massive wave of cheating, banning over 50,000 accounts from the game. This helped to restore some semblance of fairness to the game, but nonetheless, it is estimated that the cheating plague caused major losses to the brand.  

Lately it had to ban 14K accounts again> 

Moreover, in its effort to satisfy honest players and support fair play, Call of Duty now shows you when a cheater has been booted right in the kill feed. source  

Destiny 2

Another example includes this recent news of ‘Destiny 2’ developer Bungie winning $6.7M in lawsuit against video game cheat-maker. source

The article states that:  “It’s an endemic problem in online gaming, particularly on the PC platform” 

Overwatch 2

An additional example is Overwatch 2 which has banned 250,000+ cheaters since launch.


Valve banned 40,000 Dota 2 cheaters; The company said it was able to spot players that read ‘secret’ data not available during normal gameplay, as part of an ongoing campaign to reduce cheating in the game (Source

There are many more examples, and one can say that the industry as been busy fighting this pandemic for quite some time now, with only partial success. 


The cost of cheating: A multifaceted burden on the gaming industry

Cheating in online gaming creates more than just frustration—it messes with fair play, saps player excitement, increases costs, and hurts brand names. Recognizing these issues is key for game makers and industry players to tackle this widespread problem head-on. 

We can split this discussion into discrete areas: 


The impact of cheating on IAP and player engagement

Cheating directly impacts in-app purchases (IAP) by deterring legitimate players from spending money on a game if they perceive that the playing field is uneven. A study revealed that 60% of gamers across the globe have had their multiplayer gaming experience negatively impacted by other players cheating. And 77% are likely to walk away from a multiplayer online game if they feel that other players are gaining an unfair advantage through cheating. 

This translates into lost revenue for game developers, who rely on IAP to sustain their businesses – another study shares that 48% of online gamers said that they would buy less in-game content as a result of cheating

Moreover, cheating erodes player engagement and retention, as players become discouraged by unfair competition and seek out more enjoyable gaming experiences. A survey by Newzoo revealed that 57% of gamers have quit a game due to cheating, highlighting the detrimental impact on player loyalty.


The cost of cloud 

The increased traffic generated by cheating tools can overload game servers, leading to performance issues and additional costs associated with server maintenance and upgrades.


Industry estimates of the financial costs of cheating

Industry experts estimate that cheating costs the gaming industry billions of dollars annually in lost revenue, increased operational expenses, and reputation issues. For example, this resource reports that the industry lost $29 billion in 2019 alone, due to video game cheating. 


The indirect impact of cheating on different areas

Beyond the direct financial costs, cheating also has indirect consequences that affect various aspects of the gaming experience.

Competitive Gameplay: Cheating damages the gamer experience which leads to a decline in viewership and sponsorship opportunities.

Overall Gaming Experience: Cheating creates frustration among legitimate players. This leads to a decline in player engagement and increases churn. 

Game Reputation: Cheating can lead to negative publicity around the game brand, damaging brand image and future revenue potential. 


Combating cheating: The industry’s arsenal

In the battle against cheating, the gaming industry has deployed a multifaceted arsenal of countermeasures, ranging from sophisticated anti-cheat software to comprehensive cheat mitigation policies.


Anti-cheat software and technologies: The first line of defense

At the forefront of the anti-cheating fight lies a diverse array of software and technologies designed to identify and neutralize cheating attempts. These tools employ various techniques, including:

  • Signature-based detection: This method involves identifying and blocking known cheating patterns or signatures within game code.
  • Heuristic analysis: This approach analyzes player behavior and game data to detect anomalies indicative of cheating, such as superhuman reflexes or impossible movements.
  • Machine learning-based tools: Advanced algorithms are trained on vast amounts of data to identify patterns and anomalies that could signal cheating activity.
  • Kernel-level protection: Deeply embedded anti-cheat software can monitor and intercept system calls, gaining access to a lower level of the game’s operation to detect and prevent cheating attempts.

Cheat mitigation policies: Beyond detection; addressing the root causes

While anti-cheat software and technologies play a crucial role in detecting and preventing cheating, cheat mitigation policies address the underlying causes of cheating. These policies outline clear guidelines on acceptable behavior, penalties for violations, and appeal processes.

It is important to mention here that identifying cheaters doesn’t necessarily mean that companies will prevent them from playing – it depends on the impact on the game community and the game’s monetization. In the past, companies chose to either turn a blind-eye to cheating as long as their user base was growing or simply ban players for a lifetime to keep the game clean. Today, gaming companies evolved their policies but still, they need advanced anti-cheat detection tools in order to have full visibility over what’s going on in their games. For instance, if players use simple cheats to improve their own experience without impacting others, companies may decide not to ban them.  Understanding cheaters’ motivation also contributes to optimizing the game, so players will get what they want without using cheats. 

When cheaters lead to monetization loss, game publishers insis on a rapid action that stops this impact. 

By combining sophisticated anti-cheat technologies with comprehensive and dynamic cheat mitigation policies, the gaming industry can create a fairer and more enjoyable experience for all players. 


Problems and limitations: 

However, things aren’t perfect…

Most anti-cheat technologies often fall short of addressing the ever-evolving tactics of cheaters, leaving loopholes that can compromise fair gameplay. Moreover, some tools have limited scope and can’t identify new or modified cheating methods. Some tools require high maintenance. False positives are also a major concern, as they result in intolerable inaccuracies that unjustly subject innocent players to policing. Several of these tools introduce privacy-related challenges and there are other limitations not mentioned here.  


The role of machine learning: adaptive defense against evolving threats

Machine learning (ML) has emerged as a powerful tool in the anti-cheating arsenal. ML algorithms can continuously analyze vast amounts of game data, identifying patterns and anomalies that indicate cheating activity. This adaptive approach allows anti-cheat systems to stay ahead of evolving cheating methods, as ML can learn and adapt to new threats.  

But, there’s a problem, again… 

The role of machine learning in the battle against cheating is a double-edged sword. While it empowers anti-cheat systems to detect sophisticated techniques, it also provides cheaters with tools, like ChatGPT or Google Bard, to create advanced cheating software that can deal with the newest anti-cheat tools.
This leads to an ongoing arms race, requiring constant refinement of anti-cheat algorithms. 

So, what’s the deal? 


The missing element- Device Sensor-Based In-Game Analytics

As mentioned, traditional anti-cheat methods, while effective to some extent, often fall short in addressing the sophistication and adaptability of modern cheating techniques. 

Gaming devices are equipped with a plethora of sensors that capture a wealth of data about user actions and interactions with the device in real-time. These sensor signals, ranging from accelerometer and gyroscope data to touch and pressure inputs, provide the most valuable insights into player behavior and identify anomalies in a way that no other technology can. 

For instance, a sudden spike in movement speed or an unnatural precision in aiming can be indicative of the use of aimbots or wallhacks. Similarly, discrepancies between in-game actions and sensor data can reveal attempts to manipulate the game mechanics.

Sensors help us understand human behavior in its most refined/raw form, and are very hard to trick. 


The Benefits of Using Mobile Device Sensor Data for Anti-Cheat Purposes

Incorporating device sensor data into anti-cheat strategies offers several key advantages:

  • Real-time detection enabling immediate identification of suspicious behavior and cheating attempts.
  • Granular data analysis that allows a much more precise and nuanced analysis of potential cheating activities compared to all other tools. 
  • Dealing with evolving threats by identifying patterns and anomalies in real time.
  • Privacy is fully kept as the technology doesn’t include a private-data sensor such as camera, microphone and does not collect PII (personal identifying information).
  • Clear visual proof of the cheating behavior can be generated.

Addressing the everlasting ML-based arms race with sensor-based analytics:  

By focusing on real-time data from in-game sensors rather than solely relying on algorithmic detection, this approach introduces a more immediate and context-aware defense against cheating tactics. This shift minimizes the reactive nature of traditional anti-cheat methods, providing game developers with a dynamic and effective means to stay ahead in the ongoing battle against gaming misconduct.


Overcoming traditional game analytics gaps

Sensor-based analytics can identify subtle and dynamic patterns in player behavior and device interactions that event-based analytics cannot detect, enabling the recognition of sophisticated cheating methods that would otherwise go unnoticed.

For instance, event-based analytics may be able to identify a player who suddenly teleports across the map, indicating the use of a wallhack. However, it would struggle to detect a more sophisticated wallhack that dynamically adjusts its behavior to avoid detection. Sensor-based analytics, on the other hand, can continuously monitor the player’s movement and device interactions, identifying subtle changes in aiming patterns, movement speeds, and input timing that may indicate the use of such a cheat.


How does it work? A couple of examples: 

Speedhack Detection

Speedhack cheating involves using software that increases the speed of a player’s character, making them difficult to catch. Sensor-based AI solutions can detect speedhacks by analyzing data from the device’s accelerometer and gyroscope. These sensors track the movement of the device, which can reveal unnatural movement speeds that are indicative of speedhack usage. For instance, a speedhacker might exhibit an unrealistic acceleration or maintain a constant speed that exceeds the limitations of the game’s physics. Sensor-based AI algorithms can identify these patterns and flag the player for further investigation.


Recoil Control Cheating

Recoil control cheating involves using software that automatically compensates for the recoil of a weapon, making it easier to maintain accurate fire. Sensor based analysis can reveal it. For instance, a recoil control cheater might exhibit a consistent downward movement of the device, compensating for the upward recoil of the weapon. 


Rapid-Fire Cheating

Rapid-fire cheating involves using an external controller that can send multiple inputs to the game in a very short period, resulting in an abnormally high rate of fire. Sensor-based AI solutions can detect rapid-fire cheating by analyzing data from the external controller’s input events. These events include button presses, joystick movements, and other actions that are registered by the controller.


Unrealistic Aiming Assistance

Unrealistic aiming assistance cheating involves using an external controller that provides aimbot-like assistance to the player. For instance, a cheater using an unrealistic aiming assistance device might exhibit an unnatural smoothness in their aim or a constant lock-on to opponents. Sensor-based AI algorithms can analyze the timing, precision, and consistency of input events along with the movement patterns of the device to identify these anomalies and flag the player for further investigation.


Adaptive Wallhack and Aimbot

Traditional anti-cheat tools often rely on signature-based detection and heuristic analysis, which can be ineffective against sophisticated cheating methods that adapt to countermeasures. Sensor-based AI solutions, on the other hand, can continuously learn and adapt to identify new cheating techniques, including adaptive wallhack and aimbot.

Adaptive wallhack and aimbot utilize machine learning algorithms to analyze in-game data and sensor inputs, constantly adjusting their behavior to circumvent traditional detection methods. These cheats may employ techniques like dynamic aiming patterns and real-time adjustments based on player actions, making them difficult to identify using conventional approaches.

Sensor-based AI solutions can overcome these limitations by continuously monitoring player behavior, device interactions, and sensor data. AI algorithms can analyze patterns and anomalies in real time, identifying subtle changes in aiming patterns, movement speeds, and input timing that may indicate adaptive cheating.

Event-based analytics, which relies on analyzing discrete events in the game, falls short in detecting adaptive cheating methods due to its inability to capture the continuous and dynamic nature of these cheats..


Pretends to be human, fulls “regular” in-game analytics: 

For instance, an adaptive aimbot might exhibit slight variations in its aiming trajectory, appearing less predictable and more human-like to traditional anti-cheat tools based on event-based analysis. However, sensor-based AI algorithms can easily detect these subtle variations.


The Future of Online Gaming Cheating

The world of online gaming is constantly evolving, with new game mechanics, complex environments, and emerging technologies creating new opportunities for cheaters to exploit. Traditional cheating methods, such as aimbots and wallhacks, continue to pose a threat, but new techniques are emerging that are more difficult to detect and counter.


New and Emerging Cheating Methods

Machine learning-powered cheats are becoming increasingly prevalent, utilizing algorithms to analyze in-game data and adapt their behavior to avoid detection. These cheats can dynamically adjust aiming patterns, movement speeds, and input timing, making them difficult to distinguish from legitimate player actions. 

Moreover, the rise of cloud gaming and cross-platform play introduces new challenges for anti-cheat measures.


How Technology Will Evolve as Well

As a consequence, anti-cheat technologies are (and will continue to) adapt as well.  Sensor-based ML analytics, which leverages data from device sensors like accelerometers and gyroscopes, offer an example of such adaptation.


False Positives: A Critical Concern

One of the biggest challenges with cheating is the need for accuracy. False positives, where legitimate players are mistakenly identified as cheaters, can have severe consequences, causing frustration, alienating players, and damaging the game’s reputation. Emerging anti-cheat solutions will employ robust evidence-collection and analysis techniques. 

Sensor-based data and AI insights should be combined with traditional anti-cheat methods to create a comprehensive and reliable detection system.


Evidence-Based Decisions: The Key to Effective Anti-Cheat

Evidence plays a crucial role in ensuring the accuracy and fairness of anti-cheat measures. By collecting and analyzing a wide range of data, game developers can gain a deeper understanding of cheating patterns and make informed decisions about how to combat them.

This evidence-based approach allows for more targeted and effective anti-cheat measures, reducing the risk of false positives and ensuring that legitimate players are not unfairly penalized.

In conclusion, the future of online gaming cheating presents both challenges and opportunities. As cheaters become more sophisticated, anti-cheat technology must adapt and evolve, leveraging advancements in sensor-based analytics, AI, and evidence-based decision-making to maintain a fair gaming experience for all players.



Cheating in online gaming is like that annoying kid at the playground, but in the gaming world, it’s a financial buzzkill. From aimbots to cloud-based mischief, and in terms of the cost of cheating, the industry’s bleeding billions. 

Enter anti-cheat technology, AI, and sensors—the ultimate defense against cheat wizards. With this tech arsenal, fair play is more likely to prevail in the gaming universe. ??



Mitigating in-game cheating: An overview of modern anti-cheat strategies

If you’re part of the gaming industry, you’re likely well aware of the pervasive problem of in-game cheating, ranging from hacked accounts in online multiplayer games to exploits in single-player experiences. Studies suggest that nearly 70% of gamers have encountered cheating and inauthentic behavior, leading to a decline in player trust, engagement, and ultimately, revenue. The loss can be roughly sliced: The loss of revenue from in-game stores due to IAP cheating, trading digital assets that were acquired fraudulently, and weakened player retention.

Developers and publishers suffer from reputational damage, direct and indirect revenue losses, increased churn, and more, due to the continuous surge of online game cheating.

To combat this growing threat, the industry invests in ever-evolving anti-cheat strategies, such as advanced detection algorithms, stricter penalties, and community-driven initiatives. The battle against cheating is a constant arms race, demanding continuous innovation and adaptation to keep pace with the evolving tactics of cheaters.

Cheating types plaguing online Games – A few examples:

  • Automation: Scripts that automate gameplay, giving players unfair advantages such as enhanced skill level, higher win rate, or more resources by playing non-stop for longer periods.
  • Farming: Exploiting automated gameplay to gain resources or experience bypasses the intended gameplay loop (e.g., Elden ring).
  • Emulators: Running games on unauthorized platforms can potentially grant unfair advantages (e.g., Call of Duty).
  • Hot Keys: Instant execution of complex actions removes the need for manual skill
  • Input Devices/Modifiers: Rapid-fire controllers bypass human limits, offering impossible accuracy (e.g., Destini-2).
  • Wintrading: Manipulating the matchmaking system for easy wins, often for personal gain (e.g., wintrading rings in League of Legends).

Additional cheating types are related to commercial cheating (vs. indie cheating):

– Account selling
– Grey market (selling assets outside of the game and trading them in-game)
– Account takeover
– API attack/backend spoofing

How wide is the cheating phenomenon?

Surveys conducted in recent years indicate that nearly 77% of players have encountered cheaters during their gaming sessions, leading to frustration and jeopardizing the integrity of the gaming experience.

Of the online gamers, only 12% have never had their multiplayer gaming experience negatively impacted by other players’ inauthentic behavior. In other words, about 9 in 10 players have had a negative experience because of cheaters.

Not surprisingly, 76% of respondents also stated that it is important for them that multiplayer games online are secured against players with unfair advantages. Source

Moreover, it is estimated that cheating costs the industry up to $29 Billion!

Traditional anti-cheat strategies: strengths and weaknesses

Pattern-based detection

Pattern-based detection identifies known cheating patterns and behaviors. This approach is similar to how antivirus software works. It relies on pre-defined patterns of known cheats to flag suspicious activity.


  • Effective against known cheats.
  • Relatively simple to implement.


  • Reactive approach, unable to detect new cheats until they are identified and patterns are created.
  • Vulnerable to modifications of existing cheats, requiring constant updates and pattern maintenance.
  • Event Specificity: Limited to detecting cheats similar to the ones it was trained on.

Anomaly-based detection

Anomaly-based detection analyzes player behavior and game data to identify suspicious patterns that deviate from normal gameplay (Anomalies). This approach relies on algorithms that learn from known cheats and attempt to generalize to identify new ones.


  • Can detect new cheats based on suspicious behavior patterns.
  • More adaptable than pattern-based detection.


  • Can be prone to false positives, flagging legitimate players as cheaters.
  • Requires careful tuning to avoid false positives and false negatives. Hence it’s time-consuming and expensive.
  • Constant Adaptation: Requires constant updates to keep up with evolving cheats.
  • inability to detect cheats that closely mimic human behavior, as these cheats do not create anomalies. One example of such cheats is the “reply attack.”

Server-side validation

Server-side validation involves running game logic on the server side to verify the accuracy of player actions and game state. This approach aims to prevent the manipulation of in-game data by cheaters (also called moding).


  • Provides a high level of security as game logic is not exposed to client-side manipulation.
  • Can detect some sophisticated cheats that modify game data.


  • High resource consumption on servers.
  • Can introduce lag and performance issues for players.
  • Effective against specific types of cheats, particularly those that exploit server-side vulnerabilities.

Summarizing the Limitations of Traditional Anti-Cheat Strategies

While traditional anti-cheat methods play a crucial role in preventing cheating, they face significant limitations:

  • Reliance solely on limited pre-defined game events – This creates huge blindspots as it looks only at what actions that players took without the nuances that show how they take these actions- what strategies are they using (which is a key in cheating)
  • Reactive Nature: These methods require identification and response to new cheating methods as they emerge, leaving a window of vulnerability for cheaters to exploit.
  • Inconsistency in Effectiveness: The effectiveness of these methods can vary depending on the type of cheat and the skill of the cheater. Sophisticated cheaters may be easily able to bypass traditional detection methods.
  • High Resource Consumption: Server-side validation and manual moderation require significant resources from game developers and publishers, potentially impacting performance and scalability.

These limitations highlight the need for more advanced and proactive anti-cheat solutions to combat the evolving landscape of cheating in online games.

ML-based anti-cheat strategy with sensor-powered justice

As mentioned, traditional anti-cheat methods offer valuable tools, but their blindspots leave room for improvement. A new emerging approach is leveraging the power of ML and mobile device sensor data signals to combat cheating more effectively.

Sensor-based game analytics

Gaming devices come equipped with sensors like touch metrics, accelerometer, gyroscope, and more, tracking user interactions and movements. These sensors generate a unique dataset, offering insights into not just what players do, but how they do it with unprecedented granularity. Analyzed with the right tools, this exclusive dataset offers valuable insights into player behavior, uncovering hidden opportunities for anti-cheat detection.

ML-powered analysis

Machine learning algorithms analyze sensor data to identify in-game anomalies and recognize patterns that deviate from typical gameplay in real-time. This allows for the detection of suspicious behavior even if the specific cheat being used is unknown. There are plenty of products able to detect cheaters, but an ML-based solution learns from players and adapts itself to fit the specific game. Moreover, as it keeps being trained on the data, it is always kept up to date with no need for manual maintenance when new versions of the game are released or new cheats are deployed by the cheaters.

Most importantly, an ML-based solution that leverages device sensor signals allows an unprecedented understanding of human behavior, thereby providing a new level of accuracy in detecting unauthentic player behavior, including cheating. This enhanced accuracy makes its cheat detection much harder to fool.

Proactive approach

Unlike traditional methods that react to known cheats or are limited to pre-defined tracked events, AI-based anti-cheat solutions offer a proactive approach, simultaneously analyzing huge chunks of data signals.

Beyond the obvious: sensor-based analytics unsung advantages

Leveraging mobile device sensor data in conjunction with AI offers unique advantages for anti-cheat solutions:

  • Unveiling Hidden Anomalies: Sensor data captures subtle movements and interactions uncovering intricate patterns that traditional methods can’t detect, hence enabling early identification of potential cheaters.

For instance, traditional methods struggle to detect automation scripts like rapid-fire controllers, which bypass human limitations and offer impossible accuracy. These scripts automate key presses and actions, simulating superhuman reflexes and aiming precision. However, sensor-based AI can analyze subtle variations in player input, such as touch durations and micro-movements, revealing patterns indicative of automated behavior. This allows for the detection of previously undetectable cheats, ensuring a fairer and more competitive environment for all.

  • Adaptability to Evolving Tactics and Continuous Learning: Machine learning algorithms trained to analyze human behavior in diverse sensor data can adapt and evolve to recognize new cheating techniques, providing a more robust and future-proof solution. As cheaters are constantly innovating, devising new ways to exploit game mechanics and gain an unfair advantage, this benefit is crucial.

For example, imagine the emergence of a new hotkey script that allows players to instantly execute complex actions in a game similar to League of Legends. While traditional methods might not initially recognize this new form of cheating, sensor-based AI can detect subtle changes in touch patterns, such as the duration and frequency of touches, which can reveal the use of hotkey scripts. Alternatively, it can measure the time it takes for a player to react to in-game stimuli, which can help to identify automated scripts that provide superhuman reflexes.

In our hotkey script example, sensor-based AI would be able to detect the script by analyzing the changes in the player’s touch pattern. As the player presses the hotkey and the script executes the complex action, the AI observes a sudden change in the timing and sequence of touches. This deviation from normal behavior would be flagged as suspicious and could be investigated further.

Efficiency through Automation: Sensor-based AI automates cheat detection, reducing the need for manual reviews and freeing up resources for game development.

To illustrate this, let’s compare the identification processes:

Traditional anti-cheat example:

  • A player reports suspicious activity to the game developer.
  • The developer gathers evidence, such as gameplay footage and logs.
  • Analysts manually review the evidence to determine if the player is cheating.
  • If the player is found to be cheating, the developer takes action, such as banning the player from the game.

This process can be slow and inefficient, with several drawbacks:

  • Slow detection (days or weeks).
  • Limited scope.
  • Human bias leads to false positives/negatives.

Sensor-based ML solution:

  • Sensor data from a player’s device, such as touch metrics, gyroscope, and accelerometer, is collected.
  • Machine learning algorithms analyze the data in real-time to identify patterns that are indicative of cheating.
  • Once the AI detects suspicious activity, bias-free, the player is flagged for applying the relevant policy.
  • As the false-positive rate is low, down to effectively non-existent, there is no need for manual review.
  • If needed, provide flagging reasons and evidence for a quick manual review process (AI explainability).

Learn more about Sensor-based ML tech

Additional benefits include:

  • Personalized Player Profiles: AI analyzes individual, non-aggregated sensor data patterns. This allows for more accurate identification of anomalies and suspicious behavior specific to individual playstyles.
  • Scalability for Diverse Environments: Sensor-based AI solutions adapt to different game environments and player bases, making them effective for online games of varying sizes and genres.

What news does it bring for game publishers?

Sensor-based AI translates to catching cheaters within minutes (not days), achieving maximum recall, liberating developers from manual work, resulting in no false bans (happy players), and providing non-skewed clearer data insights for superior games. Developers no longer need to continually introduce new event logs to track down cheaters, adjust their anti-cheat solutions with the release of new versions, or grapple with a high false positive error rate.


As cheating in online games evolves, anti-cheat mitigation strategies are not staying behind. Emerging methods go beyond traditional detection by scrutinizing subtle player behaviors and harnessing ML engines, thus uncovering concealed patterns and adapting to fresh tactics. This not only ensures a fairer gaming environment but also liberates resources for game developers to focus on improvements and innovation, ultimately enhancing the player experience.


Online gaming churn-prediction – 2024 challenges and new methodologies

We all know that in the online gaming sphere, where player loyalty is both a coveted currency and a constant challenge, being able to accurately predict churn is an important key.

Accurate churn prediction is essential to optimizing player retention, unlocking opportunities like precise cross-promotions, tailored re-engagement incentives, and ad-based monetization that does not interfere with experience or cannibalize retention.

In a nutshell, churn prediction analyzes player behavior patterns to identify those at risk of leaving, using advanced algorithms to trigger targeted interventions like re-engagement campaigns and game design optimization, ultimately boosting player retention and profitability


Online-gaming retention – Industry benchmarks

Recently, research by AppsFlyer found an average retention rate of less than 30% on day 1 for mobile games, dropping to less than 8% on Day 7, as you can see in the chart below (These numbers are also dropping year over year as you can see):


Another research by GameAnalytics reveals a retention rate of between 2% to 20% after 7 days:

Day 30 data shows, unsurprisingly, much lower retention rates:

Do you need extra proof – this table is based on Adjust research:


As it may look harsh in the eyes of outsiders, industry professionals are familiar with these trends and are hardly surprised. However, understanding these benchmarks is important when looking at the soaring UA costs.

Retention, hence churn, directly determines whether a user shows a positive or negative ROI, making accurate churn prediction crucial for extending engagement and maximizing lifetime value.

Moreover, retaining existing players is significantly cheaper than acquiring new ones. Researches show that acquiring new users can cost 6-7x more than retaining existing ones, and can increase revenues by over 25%!


The problem with today’s churn prediction strategies:

Traditional churn prediction methods, while providing a valuable starting point, struggle to keep pace with the dynamic reality of modern online gaming. Their reliance on historical data and static player profiles creates several key limitations that result in insufficient accuracy:

1 .Data acquisition lag, model stability issues, resource limitations, engineering bottlenecks, all these challenges result in models being trained on months-old data, therefore struggling to adapt to the ever-evolving landscape of games. Major releases, feature updates, and changing player behavior can render their predictions obsolete. Retraining on fresh data becomes a constant, resource-intensive effort that impacts scalability.

2 . Newbie Blind Spot: With a typical requirement of 6-12 months of player data, these models are blind to the nuances of new players. Early churn signals are missed, leaving valuable opportunities for engagement untapped.

3 .Frustration in the Shadows: Real-time behavioral patterns, crucial indicators of frustration and impatience, often go undetected by traditional methods. Subtle changes in gameplay, interaction with features, and communication patterns can be missed, leading to inaccurate churn predictions.

4 . The Blurry Lens of Granularity: Traditional models often paint players with a broad brush, failing to identify specific cohorts or sub-segments within the churn-prone user base. This lack of granularity damages targeted interventions and personalized re-engagement strategies.
As a consequence, game publishers, including tier-1 developers that are known to have enormous resources turned to technological innovation, are openly reporting their frustration driven by the relatively low accuracy of their churn prediction capabilities.


AI-based, sensor-based – Behavioral analytics for the rescue

As discussed, traditional churn prediction methods often grapple with static player profiles and data silos.

Enter behavioral analytics, a transformative approach that leverages the power of AI and mobile device sensor signals to paint a dynamic picture of player behavior right from the first session.

By analyzing touch gestures, screen tilt, and other subtle cues, behavioral analytics unlocks deep insights into player engagement, satisfaction, and even frustration with unmatched accuracy.

Unlike methods that rely on long-term historical data or must wait for specific events to trigger predictions, behavioral analytics operates in real-time and is based on immediate physical signals that aren’t predefined by the developers like logs.

Think of it as having a window into the player’s mind, offering an immediate understanding of their emotional state and potential churn risk. This allows for timely interventions, like personalized incentives or targeted re-engagement campaigns, before players reach the point of departure.

The accuracy of AI-powered behavioral analytics is surpassing industry standards by a significant margin due to richer data sets, real-time processing, and of course the advantages that come with AI technology.

Studies have shown models trained on mobile sensor data can predict churn with much higher accuracy compared to all other tools, even within the first 24 hours of gameplay.

This opens up a world of possibilities, especially when it’s about new players, where churn prediction is no longer a guessing game.

Also, by capturing the nuances of player behavior, behavioral analytics offers a significant advantage over traditional methods that are dependent on pre-defined events. It tracks subtle shifts in engagement, pinpointing potential churners before they become lost causes and right when frustration starts, for example.

In essence, behavioral analytics isn’t just about predicting churn, but about truly understanding players’ emotional state as reflected in their behavior.


Specific advantages:

Insights as early as 24 hours from the initial session: Game professionals are struggling to gain insights from new players. Waning playtime, hesitant gestures, and frustrated taps –all these are gold-worth indicators when discovered before they escalate into full-blown churn. Sensor-based early warning system allows developers to intervene before players reach the tipping point. It can therefore predict churn for fresh players, within the first 24 hours of their journey.

Churn risk segmentation – A laser-focused approach: Gone are the days of one-size-fits-all interventions. Behavioral analytics segments players based on their churn risk and playing behavior, enabling developers to craft targeted strategies. Players on the brink receive immediate attention, while others at moderate risk benefit from personalized nudges to keep them engaged.

Engagement Optimization: By analyzing player in-game behavior within specific game features, developers can pinpoint what sparks joy and what breeds frustration in means that are much more accurate compared to event-based logs. This data-driven insight allows product teams to optimize in-game elements, balancing challenge with reward, and ultimately creating experiences that keep players hooked.


The future of churn prediction: AI at the forefront

As the gaming landscape evolves, the ability to predict and prevent player churn becomes ever more critical.

The reliance on static data and limited ability to capture real-time emotions is no longer enough. As in many other industries, AI and machine learning step into the spotlight here as well. Continuous advancements in AI and machine learning algorithms are already paving the way for more accurate and nuanced churn prediction. By leveraging advanced neural networks and deep learning techniques, these models can analyze vast amounts of data, including mobile sensor signals, in-game behavior, and even player communication patterns. This allows them to identify subtle shifts in player engagement, frustration, and satisfaction, providing early warning signs of potential churn before it becomes irreversible.

Their ability to analyze ultra granular data signals at scale and in real-time allows them to accurately recognize patterns that otherwise would have stayed hidden, just because such behavioural patterns are not limited to the human mind.

Today, the standard in the industry is rapidly changing and churn models can predict a player’s risk of leaving within hours of their first login, supporting developers, product, and acquisition teams to exceed their KPIs.

This is about deeper connections with players and getting to know them better, just like humans do with each other, fostering a more engaging and satisfying gaming experience for all.

By understanding the player’s emotions, motivations, and frustrations, developers can create games that truly resonate and keep players coming back for more. In this way, AI becomes not a tool for manipulation, but a bridge to a healthier, more sustainable gaming ecosystem.
Why it makes so much sense? Think about human beings. We all tend to “learn” other people that we interact with. What’s their style? What makes them happy? Annoyed? Impatient? We are not simply leaning on explicit communication, but also on nuanced gestures. Eye movements, body position, hand movements, and more.

That is because we all know that body language tells a story that is worth a thousand words. Behavioural analytics is harnessing AI-based device sensor data in order to do the same when humans interact with software, just like in online gaming.

The future of churn prediction and retention optimization will be about crafting games that will “read’ the player and will instantly adapt to minor behavioral changes, as reflected in the device sensor signals.



In the dynamic world of online gaming, predicting and preventing player churn is paramount for sustained success. Current churn prediction methods face challenges such as data lag and a lack of granularity, making them inaccurate and therefore prompting frustration among industry leaders.

Enter behavioral analytics—a transformative, AI-powered approach that advances churn prediction capabilities. By analyzing mobile device sensor signals, this method provides real-time insights into player emotions and behaviors, predicting churn with much higher accuracy compared to all other methods, and as fast as within the first 24 hours of gameplay.

Its advantages include early insights, precise churn risk segmentation, and data-driven optimization of in-game features.

Looking ahead, AI and machine learning take the spotlight, using advanced neural networks to decode nuanced player emotions. This shift towards predicting churn risk within hours of a player’s first login heralds a transformative era, fostering deeper connections with players and creating a healthier gaming ecosystem.





Ready to get started?
Let’s talk