As a result of digital change, data is being generated in all areas and is available for analyzing performance, trends and developments. In professional soccer, too, the use of players' performance data by coaching staff has become established. The particular challenge lies in the live analysis of game data.
For our customer, a top club in the soccer Bundesliga, we have created a high-performance tech stack based on Azure and Snowflake. It forms the core of the taod Football Data Platform. Thanks to the special data processing, the analysis can be started directly in Tableau and thus generate quick added value for coaching staff and team alike.
How does live data analysis work?
How can extremely high volumes of data be processed?
How is a cloud-based technology stack made up?
A high-performance ETL process in Python handles data retrieval, quality checks and initial transformations. Orchestrated via Logic Apps and executed in Docker containers via Azure Container Instances, it ensures consistent deployment and scaling as needed — critical for processing with minimal latency during live games.
The data platform is based on Snowflake in combination with Azure and provides a stable, scalable and low-maintenance foundation. Through multi-level data storage, different versioning can be managed. Live data and quality-assured raw data are processed separately — even peak loads with millions of events per game day are reliably handled.
dbt Cloud is used to develop and orchestrate the data models. All transformations are comprehensibly implemented in SQL, versioned and tested. Thanks to a direct connection to Snowflake, an agile workflow with integrated documentation and data lineage is created — ideal for a scalable and maintainable data model.
Interactive analyses and dashboards are carried out via Tableau Server in a self-service approach. The live connection to Snowflake enables real-time analyses that are accessible to a wide range of stakeholders via an intuitive user interface. Teams can create their own dashboards and derive data-driven decisions directly from the tool.
Football analytics is now a key success factor for sporting and economic development. With the new technology stack, the club analyses every game based on data, enriches the raw data in a context and visualizes it as animations. In this way, tactical knowledge is conveyed in an understandable way — from analysts to coaching teams.
In the project, over 3.6 million Bundesliga data points per game were retrieved in real time and stored in a database. KPIs calculated from this were visualized and made available for analysis. By setting up its data platform, the club laid the basis for an efficient analytics department—with direct added value for the team's sporting performance. The platform now enables well-founded tactical decisions, detailed opponent analyses, and targeted performance assessments of individual players.
The aim of the project was to combine live game data in a central football data platform, evaluate it in real time and make it directly usable by coaching staff, analysts and teams. The top club in the Bundesliga wanted to place data-based tactical decisions, opponent analyses and performance evaluations on a scalable technological basis.
The central challenge was to process very large amounts of live data per game with minimal latency, calculate your own KPIs and visualize the results directly. In addition, other data such as scouting, health and training data should be able to be integrated so that game analysis does not remain isolated.
taod developed a cloud-based tech stack using Microsoft Azure, Snowflake, dbt Labs and Tableau. This included a high-performance ETL process in Python, a central data platform in Snowflake, versioned and tested data models with dbt cloud, and interactive Tableau dashboards with a live connection for self-service analyses.
The introduction began with the analysis of the available data sources and the development of an initial ETL process. The central Snowflake data platform and data models were then built before the live data was visualized via Tableau.
With the new platform, the club can retrieve, store and analyse more than 3.6 million data points in real time per Bundesliga game. The live data visualization takes place with a delay of less than 5 seconds, and the platform enables individual KPIs depending on the game philosophy and player profile. As a result, tactical decisions, opponent analyses and performance evaluations became significantly more data-driven.
Let us answer your questions during a non-binding initial consultation.
From analysis to cloud implementation: We improve your data management strategically and technologically.
We'll show you how easy it is to manage your data with Snowflake.
What components does a professional tech stack for analytics consist of?
