A data and analytics professional with 5 years of experience in the online betting industry, with a strong focus on Sportsbook and Casino.
At Allwyn, he has worked across the full data journey, bridging digital interactions with transactional behavior to deliver actionable business insight.
His background includes tag management, event tracking, cloud data engineering in BigQuery and Azure Databricks, BI development in Power BI, and advanced analytics and ML use cases such as segmentation, propensity scoring, churn prediction, and lifetime value modeling.
Most organizations collect digital analytics and transactional data in parallel — but rarely together. The result is a fragmented view of the customer: behavioral signals that never inform business decisions, and transactional models blind to how customers move through digital products.
This session presents a real-world implementation of a unified data architecture that bridges this gap. Starting from a full re-instrumentation of web and app tracking via a structured data layer and tag management, the pipeline progresses through GA4, Big Query and Azure Databricks — where digital events are joined with transactional records at the customer level.
The talk covers the architectural decisions behind a scalable, uniform event schema; a multi-dimensional data model across platform, device, channel, customer / personalization segment, and registration status; and the engineering of automated reporting pipelines in Power BI.
The second half focuses on the Machine Learning opportunity this foundation unlocks: customer segmentation on combined behavioral and transactional features; propensity scoring for deposit, bet, and reactivation events; churn prediction catching early signals weeks before the last transaction; and CLV modeling enriched with digital context.
A key theme is the unregistered user — invisible to transactional systems but rich in digital intent — and how behavioral data enables scoring and nurturing this audience before they ever register.
Searching for speaker images...