Creating a personalized gaming experience
Data engineers needed to manage infrastructure
Faster data processing for downstream machine learning
Faster data preparation, allowing team to focus on data science
Sportsbet is Australia’s biggest corporate bookmaker, with over 1.2 million customers and makes 1.1 billion bets per year. With tons of data to process and analyze, Sportsbet lacked the resources to be effective. Today, they use Databricks to streamline operations and democratize data — allowing them to leverage personalized content to increase customer loyalty and revenue.
Inability to deliver real-time results to 1.2 million customers
Data underlies every decision made at Sportsbet. Their mission is to deliver a more personalized experience for every customer when they log into their website. This requires the ability to process massive volumes of real-time data to feed into their recommendation engines in order to decide what content to display on their homepage or the betting odds of a particular team in real time.
Massive data volumes with over 1.2 million customers making 25,000 bets every minute, Sportsbet struggled to serve their customers because of limited processing capacity required to make their machine learning models work as designed.
Intensive resource requirements to build and maintain reliable and performant data pipelines. This slowed data flow to the data science team for machine learning and analytics.
Their data scientist team often spent too much time preparing and preprocessing data on their local machine, often taking months to go from exploration to production of new models.
Automated data pipelines enable downstream machine learning
With the Databricks Data Intelligence Platform and AWS, Sportsbet built a real-time personalization engine to deliver unique content for every customer with machine learning models that can be quickly improved. The data scientists and data engineer teams can focus on the work that matters the most instead of mundane tasks.
Fully managed platform with automated cluster management simplifies the infrastructure and operations at any scale.
Unified platform has allowed data science and engineering to work seamlessly with the data and each other.
Collaborative notebook environment with support for multiple languages (SQL, Scala, Python, R) enables a diverse team of users to work together in their preferred language.
Databricks naturally integrate into their AWS for a data lake approach to meet data storage, processing and warehousing needs.
Large scale data processing and ML results in faster time-to-market
With Databricks, Sportsbet is now able to do large-scale real-time data processing and machine learning much faster. The data scientist team can collect feedback on machine learning models faster from their customers, iterate and operationalize the model and and ultimately drive more innovation, better customer loyalty and increase revenue.
Improved operational efficiency: Features such as auto-scaling clusters has improved data engineering operations, with 0 data engineering time spent on managing the infrastructure.
Faster time-to-market: Data scientists can focus on actual data science now with data gathering and preprocessing at ease, 5X faster data preparation than before.
Faster time-to-insight: We now get in over a 20x performance benefit over open-source Spark with Databricks and 90% increase in time to market.
