Data Driven Spannung: Operating a Scalable Data Infrastructure

Being a market leader means we always need to stay ahead. Our data team is a strategic pillar in our company pioneering artificial intelligence concepts and cutting-edge data warehousing techniques, processing millions of transactions every day in low latency. At Tipico, we process terabytes of data every day, related to all areas of the business – such as bookmaking data, web analytics, CRM data and much more.

Over the past two years we’ve been working towards having more real-time systems while simultaneously setting up our infrastructure for the new expansion: the US side of the business. The company recently expanded to serve the US market – this brought with it the need to tailor our data processing to new regulatory requirements, as well as upgrading our capabilities to serve our latest cutting-edge proprietary sportsbook platform.

In order to achieve this, we set a goal of delivering real-time and batch data analytics within a new regulatory framework, while planning for a fast expansion. This required us to think outside the box and look towards new cutting-edge technologies. These challenges create new possibilities to innovate and enables everyone to progress and learn new skills.

We achieved this by building:

• A real-time data-lake, capable of storing data securely and efficiently, enabling users to access data which is only a few seconds old. The data lake was built using Amazon S3 and Apache NiFi. Within our flows, we enriched incoming data with additional data in order to provide a holistic view to the business users.
• To facilitate regulatory reporting, we worked with Amazon Redshift and Talend, operating intra-day loading mechanisms.
• All our existing infrastructure is in the cloud, but since the Tipico US is a fast-growing and ambitious project, we automated all infrastructure deployment to facilitate scaling and faster deployments. To do this, we used Jenkins and Terraform; coupling the “data” aspect of our work alongside popular “DevOps” methodologies.
• We provided access to data via an API Layer, along our dashboarding tool (Domo), and daily reporting (Excel) – this allowed not only business users to access our data, but also internal tooling – ensuring that other internal tools have complete data.
• Our customers are our main focus, the centre of everything we do. We set up CRM and customer interaction flows from scratch, tailored to the US market and its needs.

The project was challenging for a variety of reasons, but these challenges are what push us to go further and be better. These challenges let us Maximise Spannung every chance we get. Our attention was focused on maintaining multiple technology stacks across different time zones, all dealing with high data volumes. Whilst also, maintaining data governance and KPI definitions across two different markets. Throughout, we also ensured that our business users have all the data they need, when they need it, as the product was launching.

Tipico is continuously becoming more data driven, making the team pivotal to Tipico’s strategy.
Our focus is on developing our skills and gaining insights that will change the way we look at sports betting and data science while being extraordinary and driving innovation.