Premier League drives deep fan connection with Microsoft Foundry and Azure Cosmos DB

Turning data into a personal connection

For more than 1.8 billion fans around the world, football isn’t just a game—it’s identity, history, and belonging. Every chant, every highlight, every last-minute goal adds to a shared story that connects people across languages and time zones. The Premier League sits at the center of that story, uniting billions through competition and community.

But even the most passionate global following brings new expectations. Fans no longer simply watch; they participate, post, and personalize. They expect experiences that know who they are and what they love—moments that feel made for them. For the Premier League, this sparked a challenge to create authentic global connections through tech innovation: how do you transform decades of data, scattered across longstanding systems and partner platforms, into personalized experiences that deepen loyalty for every fan?

The answer began not with a new app, but with a new principle: No two fans are alike, and neither are the experiences that inspire them.

The Premier League is among the world’s most recognized sports organizations. Yet even with its global reach, only a proportion of its 1.8 billion fans engaged directly through the League’s owned digital channels. Fans wanted more. More immediacy, more relevance, more of a personalized connection to their game.

That meant the League needed to move beyond delivering the same information to everyone and start shaping experiences that adapt to each individual.

Behind the scenes, this required a shift from thinking about fans in broad-brush categories to thinking about them as individuals. And bringing together the right raw materials—match statistics, historical archives, editorial content, videos—to meet the needs of those individuals.

The League wanted to change, not just by upgrading its technology but by reimagining how data could bring fans closer to the game.

“We know that personalization has become increasingly important to Premier League fans with their interests more varied than ever before. Now, we are able to create personalized digital experiences which put each fan in control of the clubs, players, matches and stories most relevant to them,” says Alexandra Willis, Director of Digital Media and Audience Development at the Premier League.

To make that possible, the League needed more than infrastructure. It needed an intelligent foundation that was capable of unifying data, applying AI-driven insights in real time, and doing so with the transparency and compliance required of a global brand.

Rewriting the playbook

The Premier League’s vision called for a platform that could unify massive, heterogeneous datasets—from 30 seasons of match data to 300,000 editorial articles and 9,000 videos—into a single, queryable environment. The system would need to refresh continuously during live matches, when fan demand for personalized, contextually relevant content surges.

The League also wanted to create a multi-agent AI system that could personalize content dynamically. It wanted to pair real-time data about matches, players, and fan behavior with agentic AI to deliver stories, stats, and insights that feel personally relevant without compromising privacy.

Security and compliance were essential. As a global organization operating under regulations such as the General Data Protection Regulation (GDPR), the League needed to ensure that no personal data was stored or processed without consent. Every AI interaction would need guardrails for safety, traceability, and governance.

“We saw an opportunity to redevelop, reimagine our technology stack, and within that, our owned and operated channels,” recalls Will Brass, Chief Commercial Officer for the Premier League. “Very early in that process, we identified Microsoft as a dream partner to help us achieve that goal—a world leader in cloud services and generative AI capable of supporting what we were trying to do.”

With that partnership, the League set out to build something new that wasn’t just a digital platform, but a living system capable of understanding context, conversation, and fan emotion at scale.

Building on Microsoft Foundry

The new platform, anchored by Microsoft Azure and Microsoft Foundry, brought together cloud-native scalability with intelligent orchestration.

Using Azure Data Factory and Azure Databricks, the League ingests and processes data from more than 10 APIs, spanning 10,000 matches; 51 clubs; 10,000 players; 300,000 articles; 270,000 photos; and 9,000 videos. This data is transformed, normalized, and stored in Azure Cosmos DB, which acts as the central hub for structured match and player data, AI prompts, and anonymized conversation logs.

Taking advantage of the global distribution and low-latency design of Azure Cosmos DB, the League is working to enable delivery of personalized content in near real time, even during peak match moments. Acting as a bridge between analytics and operational AI, Azure Cosmos DB makes insights quickly available to applications and agents, reducing complexity and accelerating the delivery of differentiated experiences to users. Its flexible, JSON-based schema makes it ideal for managing diverse data types—from structured statistics to rich media—while maintaining compliance and governance across regions.

On top of this foundation, the League built agents fueled by Foundry. Within Foundry, Azure OpenAI in Foundry Models gives agents the ability to summarize, reason, and generate contextually relevant fan content. The agents retrieve data using Azure AI Search for agentic retrieval-augmented generation (RAG), helping fans and editorial teams instantly surface relevant articles, match data, or video highlights. Within these agents, plug-ins enable the Premier League Companion to:

• Surface and summarize stories from the League’s editorial archive.
• Retrieve real-time match data and player metrics.
• Generate natural-language responses, helping fans explore content through chat-like prompts.

The agents are orchestrated and connected with Semantic Kernel. This framework helps ensure that each agent interacts intelligently with others and with the League’s broader ecosystem of partner systems, including WSC Sports and Adobe.

Azure Machine Learning runs scheduled tasks that prepare answers in advance for certain questions and manages the data flow needed for those answers. These tasks use prebuilt AI models from Azure OpenAI to do the heavy lifting. Foundry provides traceability to enable a clear understanding of how each response was generated, including the flow of conversation, the questions and answers, and the data sources involved. Meanwhile, Azure Managed Redis helps maintain sub-500-millisecond latency for live interactions, caching the most frequently requested insights.

The result is a platform that doesn’t just store or deliver content—it understands the context around it.

Engineering for trust and resilience

At the scale of the Premier League, performance and privacy carry equal weight. Every component of the platform was built with security at its core.

Weekly system health checks, automated patching, and continuous monitoring help protect both fan data and application uptime. The League’s teams conduct regular red teaming exercises to stress-test open prompting features and mitigate potential misuse.

To help ensure traceability, every data interaction—from API ingestion to AI output—is logged and auditable, maintaining full data lineage and accountability. Fine-grained access controls define permissions across internal and external partners so that only authorized users interact with sensitive systems.

By embedding compliance into the design, the League created an architecture that scales more securely, balancing personalization with privacy—a crucial distinction for a global organization operating in over 200 territories.

Transforming stats into stories

The impact was immediate, with the Premier League Companion driving individualized content delivery to millions of fans. Architected to handle more than 8,000 transactions per second, it now supports millions of concurrent users, with each receiving a uniquely tailored feed. Early in the first season after launch, consumption of the Premier League app and website was up approximately 20%, year over year.

Even at peak match times, the system delivers near-instant responses, keeping fans connected without friction. And as personalization deepened, engagement soared.

“On our digital channels, we’re seeing billions of engagements. We have already had 60 million users across our website and app so far this season. Fans all over the globe are positively engaging with the content being provided to them and enjoying the stories that come off of Premier League pitches up and down the country each match round,” Brass reports.

The impact reaches beyond engagement. Internally, unified data and automated workflows have streamlined decision-making and accelerated time to market for new features. Developers now iterate faster, using shared pipelines and AI-assisted workflows to build, test, and deploy new capabilities.

Soon, individual clubs will have the opportunity to use quantifiable fan data to shape marketing campaigns, measure reach, and open new monetization channels. The League’s ability to personalize at scale has turned every fan interaction into an opportunity for connection—and every insight into a new business advantage.

Playing the long game

The Premier League’s transformation is not a one-time modernization; it’s a model for continuous reinvention. Built on Foundry, the platform evolves with each new capability, each fan interaction, and each season’s data.

The League’s roadmap includes expanded use of open-ended prompts, deeper integration of generative AI into fan and editorial workflows, along with expansion into regionalized and multilingual experiences to deepen connections with its global audience. Features like vertical video and fantasy integrations are already in development. Each enhancement builds on the same foundation: a unified, intelligent, compliant architecture capable of scaling globally while adapting locally.

By addressing the industry’s biggest challenges—fragmentation, personalization, compliance, and innovation—the Premier League has redefined its own digital experience. In doing so, it has also set a new standard for fan engagement across industries.

"More than 1.8 billion fans in 189 countries engage with the Premier League. The Premier League Companion is a storytelling platform, and the stories that we tell are enjoyed by fans all over the world every week. They are looking for the right data and insights which deepen their understanding of what they are watching, which only adds to their enjoyment of the Premier League.," Brass explains.

For IT leaders, the Premier League’s story offers a familiar parallel. Every organization—whether serving fans, customers, or citizens—faces the same challenge: connecting fragmented systems, understanding context, and personalizing experiences at scale.

With Foundry and Azure Cosmos DB, the League has built a foundation where every touchpoint becomes an opportunity to engage, to learn, and to act. Real-time insights drive better experiences, and intelligent orchestration helps ensure systems stay agile and secure. Agentic AI models turn data into dialogue, making personalization both scalable and responsible.

The result is not only a better fan app, but a blueprint for how any enterprise can use AI to get closer to its customers.

Whether leading a global brand or following a favorite club, the goal is the same: to make every connection meaningful. That’s because it’s not just data. It’s personal.

Discover more about Premier League on Facebook, Instagram, X/Twitter, and YouTube.