The Challenge

​

We were approached by an iGaming operator with the task of making the best use of their featured games section on their homepage with the goal of increasing player time on site. The operator's objective was clear: increase average player session lengths, leading directly to enhanced player engagement, satisfaction, and ultimately, profitability.

​

Our Approach

 

We sought to predict which games from their current catalogue would result in the highest average session length, based on a number of factors such as volatility and return-to-player (RTP).

​

We received daily data for 120 slot games over a 45-day period, including metrics such as Volatility, RTP, and Average Session Length. Leveraging econometric and predictive analysis techniques, we structured our analysis into clear, actionable steps:

​

    1. Data Preparation and Feature Engineering:

• Structured and prepared their data which was received in multiple Excel files to be standardised and usable for analysis.

• Identified relevant game attributes (Volatility and RTP) and temporal variables (day of week).

• Transformed categorical day-of-week data into numerical format to enhance predictive accuracy.

 

    2. Predictive Modelling:

• Implemented Regression models to accurately predict Average Session Length based on Volatility, RTP, and Day-of-Week patterns.

• Achieved a decently high predictive accuracy, demonstrating the reliability of our model.

​

    3. Actionable Insights via Visualisation:

• Clearly identified and visualised the top-performing games predicted to yield the highest session lengths.

• Provided recommendations backed by intuitive visual rankings to simplify decision-making for homepage placements.