AWS re:Invent 2025 - Driving Fan Engagement with Data, Analytics, & AI (SPF308)

Driving Fan Engagement with Data, Analytics, and AI

Challenges in the Sports and Entertainment Industry

• Consumers today have a wide variety of content and entertainment options competing for their time and attention
• Organizations in sports, media, and entertainment need to find ways to create engaging and personalized experiences to retain digital consumers

Importance of the Fan Experience

• Fans expect:
  • Exciting and thrilling content on broadcast and TV
  • Streamlined, accessible, and interactive digital experiences
  • Personalized content and recommendations
• Capturing fan data across various touchpoints is crucial to understanding and engaging with them

Building a Unified Fan Data Platform

• The NFL created a unified fan data platform on AWS that:
  • Processes over 90 billion rows of data with 250+ dimensions
  • Stores and manages over 70 million fan data profiles in Amazon Redshift
  • Processes over 1,000 data feeds daily using AWS Glue and Amazon Kinesis
  • Resulted in a 4x better understanding of fans and 2-3x increase in opt-in rates

Personalization through Machine Learning

• The Bundesliga used Amazon Personalize to personalize their app experience:
  • Analyzed 1.4 million fan preference combinations
  • Saw a 17% increase in time spent in the app and a 32% increase in article views due to personalized recommendations

Creating AI-Powered Fan Experiences

• The PGA Tour created "Torcast", an AI-powered commentary system that:
  • Processes 53 million data points per tournament weekend in DynamoDB
  • Uses Amazon Bedrock to generate AI commentary for 100% of player swings in under 10 seconds

Building a Fan Data and Analytics Platform on AWS

• Key layers of a fan data platform:
  • Data sources: Ticket sales, subscriptions, merchandise, etc.
  • Data ingestion: Using services like AWS Glue, Kinesis, Lambda, and Batch
  • Data processing and transformation: Leveraging AWS Glue, Amazon Redshift, and Amazon SageMaker
  • Identity resolution: Using AWS Identity Resolution service or EMR-based algorithms
  • AI and ML capabilities: Utilizing Amazon SageMaker and Amazon Bedrock
  • Data exposure and activation: Through APIs, BI tools, and AWS Clean Rooms

Choosing the Right Data Architecture

• Hybrid approach combining data lake and data warehouse:
  • Data lake for unstructured data, experimentation, and open-source flexibility
  • Data warehouse for structured data, BI, and known analytics requirements
  • Amazon Redshift Spectrum enables a flexible, hybrid approach

Enhancing Fan Experiences with AI

• Leveraging AWS AI services like Amazon Bedrock, SageMaker, and Quick Suite to:
  • Create AI-powered search, localization, and highlight generation for content platforms
  • Implement hyperpersonalization and virtual assistants in digital products
  • Enable interactive fan experiences and AI-powered tools for broadcast partners

Key Takeaways

• Unified fan data platforms provide a 360-degree view of fans, enabling better personalization and engagement
• Machine learning and AI can power new fan experiences, such as AI-generated commentary and personalized recommendations
• A hybrid data architecture combining data lake and data warehouse approaches offers flexibility and scalability
• AWS provides a comprehensive AI stack to build innovative, AI-powered fan experiences and solutions