SQL Pipelines & Player Metrics After modeling churn, it’s time to structure the data pipeline that feeds all metrics — just like in a real iGaming environment. Main Goals of the day Load ...

Predicting Churn with Logistic Regression Now that we’ve confirmed retention uplift, let’s try to predict churn to understand which factors drive player loss. Main Goals of the day Train a Lo...

A/B Testing Retention After generating our synthetic dataset, it’s time to test if the new feature (treatment) truly improves player retention. Main Goals of the day Compute retention rate fo...

Simulating 70,000 Players The next step in the iGaming analytics project was to create realistic synthetic data to analyze retention and churn — even without access to real players. Main Goals of...

From Players to Patterns I’ve started a new piece of the iGaming analytics project — focused on player retention and promo testing. Main Goals of the day Build an A/B test simulation on playe...

End: from Week to Month Analysis I opened Tableau for the first time, and I didn’t drag and drop. Main Goals of the day: Show 4-week retention decay, not just Day 7 Make it so clear, even a...

Dataset Improvements I scaled from 10,000 → 40,000 sessions, because 7 days don’t reveal retention patterns. Main Goals of the day: Scale simulation from 1 week → 4 weeks (28 days) Keep the...

Data visualization with Tableau I opened Tableau for the first time. MADNESS!! Main Goals of the day: Build a dashboard with 5 key visuals Show: retention, revenue, session length, bonus im...

Churn Analysis with Matchine Learning I built a churn predictor using Logistic Regression from the skit-learn library package, trained on 1,200 players. Main Goals of the day: Predict who wil...

Plotting Retention Heatmap I rewrote the cohort analysis — in Python — using Pandas, Numpy, and the simulated weekly dataset in .csv format. The choice to select Python to continue with the script...