Bundesliga transforms data into AI-powered fan-favorite stories with AWS

Commentators delivering live commentary for football don’t just describe what’s happening on the pitch—they put the action into context with data. But as the Bundesliga's data landscape intensified with event data, positional tracking, and advanced stats, the volume of data left editorial teams and commentators with an overwhelming task. 

The Bundesliga needed a way to instantly identify and surface the most relevant stories in real time while managing pre- and post-match analysis. And the challenge was not simply one of processing speed—it was architectural. Traditional data architecture approaches couldn’t handle the volume, velocity, and variety of live match data while simultaneously supporting the AI/ML capabilities required to surface compelling narratives. Bundesliga needed a better data foundation—one built for open interoperability, industrial-scale reliability, and AI-ready data quality.

AWS enabled Data Story Finder 2.0 by applying three core data foundation capabilities—open architecture, foundational excellence, and trusted data for AI—across its full solution architecture. From ingestion through AI-powered story generation and real-time distribution, AWS purpose-built data services gave Bundesliga the infrastructure to transform raw match data into 2,500 real-time narratives per season. 

At the ingestion layer, Amazon Managed Streaming for Apache Kafka (Amazon MSK) handles the continuous stream of 200 million data points generated per match. Its full wire-protocol compatibility with open source Apache Kafka allows Bundesliga to retain the flexibility of an open standard, avoiding vendor lock-in while benefiting from a fully managed service. Open architecture across the stack also enables seamless data flow between ingestion, processing, and AI services without complex extract, transform, and load (ETL) pipelines or proprietary connectors.

At the processing layer, Amazon DynamoDB, AWS Lambda, and Amazon EventBridge deliver the performance for live broadcasting demands. DynamoDB responds to hundreds of millions of requests per second with single-digit millisecond latency, which is critical when multiple goals are scored simultaneously across matches. EventBridge processes each of Data Story Finder 2.0's three story categories as fully independent, serverless workflows, so an issue with one never disrupts the others. Lambda executes story generation logic without infrastructure management, letting the Bundesliga team focus on editorial innovation. The 19+ years of AWS operational experience and consistently delivering the lowest total downtime among all cloud providers made AWS the right partner for a use case where failure during a live broadcast is not an option.

It’s imperative that the data feeding Bundesliga's machine learning models is complete, timely, and accurate. When a player breaks the season speed record mid-match, Data Story Finder 2.0 instantly identifies the achievement and delivers full context to commentators before the crowd has finished reacting. AWS's data foundation ensures trusted data for AI in practice.

When Harry Kane broke the record for the fastest shot of the 2025/2026 season against Stuttgart, he received the ball on the midfield line, drove toward the goal, and from 24.8 meters (81.3 ft) fired home at 128 km/h (79.5 mph)—with a goal probability of just 3%. Data Story Finder 2.0 surfaced the moment instantly: one of the 50 most unlikely goals and the fastest successful shot of the season.

With AWS's data foundation in place, Bundesliga generates these AI-powered stories automatically—event-based narratives triggered by significant moments, exceptional performance detection surfacing records and personal bests, and match summaries providing historical context at key moments. Commentators receive storylines like Kane's through the Commentary Live System, while editorial teams have achieved 20% efficiency gains using the Story Monitor to review or modify content before it reaches audiences. Fan-facing live tickers and official league websites now deliver data-driven stories to millions of global fans in German and English.

Bundesliga's experience demonstrates a broader truth: the quality of AI output depends entirely on the quality of the foundation. The open architecture, foundational excellence, and trusted data capabilities of AWS aren't features layered on top of a standard cloud platform—they're the reason Data Story Finder 2.0 works.