AI and Machine Learning in Gaming – Innovative Solutions & New Opportunities
Mobile games with millions of players can quickly generate vast amounts of game data. These companies require scalable, cost-effective data analytics with quick response times. To fullfill these needs, many companies use cloud services (Google Cloud or AWS) or build their own data clusters. They use advanced web-scale analytics to create personalised player experiences, using in-game telemetry and machine learning to understand player behaviour and engagement. This information helps answer questions such as:

  • At which level do players get stuck?
  • What virtual goods do they buy?
  • What actions prompt purchases?
  • Who is likely to stop playing?
  • How to appeal to both casual and hardcore players?
  • Who will spend money in the future?
  • How are game interaction and payments related?

Mobile game publisher business challenges
The mobile game publishing industry faces unique challenges due to the sheer volume of game and player data generated by mobile games. With a large number of client devices, various in-game interactions, purchases, and transactions, the analysis of mobile game data can be complex. Our client, a mobile gaming publisher and developer, serves a diverse global audience with over 500 employees and 120k new users obtained from multiple channels such as Facebook, Google, and organic sources. This results in a vast amount of customer data from all around the world.

Efficient Advertising and Customer Acquisition Costs
The correlation between customer advertisement expenses and the number of newly acquired customers was strong. The simpler the rule, the more money needed for advertising to acquire and retain more users. Our client's data analytics and performance marketing departments faced challenges in optimising marketing expenses, processing data, and storing it cost-effectively. The cost of acquiring new customers was high and only a limited number of customers made in-app purchases. Hence, it's crucial to quickly identify users who will make future purchases.

Improving Customer Retention
Within 30 days of app installation, 90% of users ceased using the app, despite significant investment in user acquisition. This high rate of lost paying users highlights the need to address customer retention issues, which will optimise marketing expenses and help retain more paying users.

Personalised User Preferences and Experience
Every user has their own preferences and each person plays differently; some prefer to quickly progress, while others take a more relaxed approach. As a result, it's crucial to have an algorithm that can predict each individual player's behaviour.

Challenges in Mobile Gaming Analytics
As our customer's app attracts more players, there is an increase in the number of users who upgrade to paid accounts and make purchases or refer the app to friends. Furthermore, various key performance indicators (KPIs) provide valuable insight into player behaviour and in-game interactions. These insights, derived from big data sets, can be used to enhance the game and improve the user experience. By utilising Game Analytics solutions, it's possible to expand gameplay time and optimise the overall user experience.

KPI’s to measure
To assess the success of a game, our customer must track key metrics such as monthly active users, monthly user acquisition, and paying users within the game. The use of KPIs, such as these examples, is crucial for measuring success.

  • MAU (Monthly Acquired Users),
  • DAU (Daily Active Users),
  • ARPU (Average Revenue Per User)
  • IAP (In-App Purchases)
  • CAC (Customer Acquisition Costs)
  • MR (Monthly Revenue)
  • YoY (Year over Year)
  • and many other
Above measures can help to understand the performance and success reasons of your game and acquired users quality.

Our solution for boosting revenue in mobile gaming combines analytics, AI, and big data.
After a collaborative workshop with both business and technical teams, we devised a plan to effectively utilise their data. Our data engineering team then established a Data Lake and Data Warehouse to collect and format all available customer information, including in-game activity such as progress, likes, interactions, preferences, expenses, demographics, and behaviour patterns. By tracking these data points, we aim to provide valuable insights into customer behaviour and preferences.

The mobile gaming analytics goal is a data monetisation
The objective of our mobile gaming analytics solution is to monetise data and enhance the gaming experience. Our solution is designed to identify crucial correlations and dependencies among large amounts of transactional data, allowing game publishers to profit from this information and improve the experience for players.

Machine Learning in mobile gaming analytics
Our mobile gaming analytics solution incorporates machine learning to analyse vast amounts of data and understand user behaviour. The machine learning models perform an intelligent analysis of each user, identifying patterns in the data and applying them to millions of users. The ML algorithms accelerate the monetisation process by quickly analysing customer behaviour and predicting future trends, while continuously adapting to changing behaviour patterns through ongoing learning with new data.

Fraud detection in player acquisition
Our team of data scientists has developed fraud detection models utilising machine learning to identify bots. These models help reduce customer acquisition costs by focusing only on high-quality customers. The solution generates automated reports on fraudulent bot activities and streams, which are delivered directly to your email inbox.

Customer lifetime value prediction (CLTV or LTV)
Our team has successfully deployed an LTV prediction AI model into production and integrated it with marketing and advertising systems. The LTV machine learning models predict the amount of money a customer is expected to spend on the application, allowing for effective segmentation of new users based on this information and customer acquisition cost (CAC). This enables our customers to selectively target those clients that are likely to generate significant revenue, saving advertising expenses on those who are unlikely to make purchases.

Customer churn prediction
Customer churn prediction machine learning models were designed to predict which paying customers are at risk of stopping their use of the application. Having this information in a timely manner enables our customers to take action to retain more users. The greatest benefit to our clients was achieved when we integrated the churn prediction system with marketing automation software and CRM systems. This allows for the automatic sending of personalised messages and tailored offers to each customer.

Recommendation engine using machine learning
The AI Group team implemented a recommendation system to boost in-app purchases (IAP). The system personalises offers and content for each player, suggesting similar games they might be interested in or offering additional purchases to help them reach their goals in the game faster. This approach improves customer satisfaction and directly contributes to higher business revenues.

Mobile gaming analytics – modern data warehouse and BI implementation
Our Mobile Gaming Analytics solution combines the latest in data warehousing and business intelligence (BI) technologies. By integrating machine learning results with the original data in a data warehouse, we enabled powerful multidimensional data analysis. The flexible self-service BI solution provided easy-to-use dashboards and alerts, making it possible for anyone to quickly create custom reports without needing any programming knowledge. Insights can be easily shared within the company. The Marketing Performance and User Acquisition departments can now analyze at-risk users, predict future revenues, and develop automated pipelines for targeting the right customers.

Results and impact of mobile gaming analytics solutions
The convergence of gaming and big data analytics is leading to significant improvements in the gaming industry. Companies such as Electronic Arts have been able to increase their advertising revenue, enhance gameplay, and better manage the user experience as a result.
Our Analytics team has empowered our customers to predict customer behavior and make informed business decisions, preventing customer churn and increasing game monetization through personalized ad offers. Predictive models and analytics systems are now the basis for key business decisions in the gaming industry.

  • Develop a sustainable and robust strategy for customer retention and acquisition
  • Formulate plans to reacquire customers who have been left
  • Convert low-revenue earning customers into highly profitable ones
  • Reduce customer defections and improve profits
  • Track customer satisfaction by product, segment, and cost to serve
  • Increase IAP purchases and user experience
Each above-mentioned activity and interaction requires AI analysis. That is why it is worth monitoring how retention and LTV changes over time, what group of users are the riskiest and how users monetisation goes. Likewise, it is also important to get alerts on such events.

In conclusion, if you have any questions regarding the above project, implementation or results just book a call and we will be glad to tell you more about benefits and deployment.