Bot Fraud in Apps: The solution is literally obvious

Gaming has always benefited from the competition among players. Whether hunting for the highest score in the arcade games of the ’80s or losing to an invincible opponent on Fortnite: being beaten by an exceptionally strong rival is not uncommon, but players and developers are now facing a new, automated threat opposite, the in-app bot fraud.

This method of fraud is already ubiquitous in e-commerce: for example, bots quickly buy limited-edition sneakers before real customers can react. But the bots are also targeting the billion-dollar mobile gaming industry. With industry sales estimated at $ 61.3 billion in 2018, that’s not surprising.

The impact affects users and developers in many ways

Bots are programmed specifically for a game to solve tasks quickly and efficiently and mimic human behavior in a sophisticated way. They behave the same way as a real player and do not trigger any security alerts. So they can freely accumulate game currency, unlock rewards and achieve seemingly unattainable rankings in the rankings. The impact of in-app bot fraud is far-reaching and affects users and developers in a variety of ways.


The success of a mobile game depends more than any other platform on the satisfaction of its users. Mobile gaming is a social experience, whether it’s a sense of community, talking to friends or competing with other players.

You might also be interested in this: The most important findings on mobile gaming

Especially in mid-core gaming, dissatisfaction is a big problem

When bots receive all awards and rewards, it is demotivating for real players trying to progress in a game. The mobile games market is hotly contested and players can currently choose from over 800,000 different games in iOS and Android app stores. If you are dissatisfied with a game, you can easily switch to a competitor.

This dissatisfaction is a big problem, especially in so-called mid-core gaming. Because here players often buy resources instead of waiting for the allocation of energy and currency in the game, which are released at certain intervals.

The generated revenue in this particular industry can be very unevenly distributed, with a small percentage of dedicated players often generating most of the revenue.

So it is the players who spend hundreds or even thousands of dollars, who care most about the perceived fairness of a game. And if they get the same results from their bets (and their sales) as someone who cheats with bots, they’ll immediately quit the game and be lost to the game as high-volume players.

You might also be interested in this: Is the future for gamers in the cloud?

Players spend money on bots instead of in-app purchases

But bots are not just a threat to usability. You can fundamentally damage the game’s revenue model. Some players use a bot subscription service for gaming bots that can be bought online without much effort to collect rewards.

So instead of making in-app purchases that are reliant on many games, players spend the money on bots, which distorts the economic structure of the game and causes sales slumps for developers.

All these threats have in common that the crux of the problem lies in the difficulty of detecting them. Exact simulation of human behavior through the use of advanced technologies such as image recognition and other machine learning techniques makes it difficult to distinguish bots from real users.

The effort required to combat bot fraud can keep developers from major game improvements and necessary updates, and further affect the usability of the game.

Such a sophisticated opponent therefore requires an equally refined reaction. And here the solution is literally obvious: records that the bot can not emulate can be used in an innovative way.

You might also be interested in this: From hobby project to game world domination

In a complex method is the solution

The key lies in the data points, which are difficult to imitate by bots. Anonymized sensor data such as touch strength, typing speed and scroll action are unique to the human user. These data can be used to develop a machine learning model to distinguish real players from bots.

This method is very complex and difficult to integrate into internal processes, but this is where the solution lies. Developing bespoke models tailored to the app’s unique user base is the only way to win against bots.

Bots have gone through a tremendous development, but they still can not mimic the unique behavior of a human user. A mobile game is growing and thriving with the use and fun of the players, so it’s a great fit that the answer to protecting these games from bot threats can literally be in the hands of the players.

Paul H. Müller is CTO and co-founder of Adjust and responsible for building and scaling the company’s technology.

Lead 4 Cta 1200X1200 V3 July 003

Women wanted in IT!

For anyone with an IT affinity, the chances of getting in the door were never better! In the current LEAD Bookazine 2/2019 we have found networks, universities and initiatives offering practical education and training. Especially women are in demand, because 55 percent of employers want to increase the female share of their IT professionals, according to Bitkom!

Now browse!

Leave a Reply

Your email address will not be published. Required fields are marked *