Validating the Premier League’s status with generative AI and fan sentiment analysis

  • Customer
  • Industry
  • Service
  • Segment
  • Author
  • Premier League
  • Media & Entertainment
  • AI/ML
  • Enterprise
  • Jodie Rhodes
Explore more
At a glance

The Premier League, celebrated as the most competitive and compelling football league in the world, oversees England’s top clubs and has a loyal global fan base. 

Challenge

The Premier League aimed to validate its status as the most competitive and compelling sports league by analysing global fan sentiment using advanced data and AI insights. 

Solution

Firemind used Amazon Bedrock’s LLM’s on AWS to analyse global fan sentiment from 12,000 social media posts on X (formerly Twitter). 

Services used
  • Amazon Bedrock 
  • AWS Lambda 
  • Amazon DynamoDB 
  • Amazon S3 
Outcomes
  • Global Validation: Confirmed Premier League’s global appeal. 
  • Out of the box thinking: Produced new thought lines for league analysis.
Business challenges

Validating the Premier League’s global appeal

The Premier League is widely regarded as the most competitive and compelling football league in the world. This reputation is built on its global reach, record-breaking broadcast revenue, and the world-class talent it attracts. However, the challenge lay in quantifying this appeal in a way that reflects not just traditional metrics, but also the emotional and psychological connections that fans feel toward the league. 

Fans generate massive amounts of content on platforms like X (formerly Twitter), and the Premier League wanted to analyse this unstructured data to uncover meaningful insights. With 12,000 tweets analysed, the project aimed to understand how fan sentiment and engagement shape perceptions of the league. This was critical to confirming its position as the world’s leading football brand. 

To achieve this, the Premier League partnered with Firemind, who introduced generative AI to analyse fan-generated content. The project used advanced tools to categorise sentiment, measure engagement, and identify themes while filtering out irrelevant data, ensuring high-quality insights. 

What our customers say

Hear directly from those who’ve experienced our services. Discover how we’ve made a difference for our clients.

Solution

Using generative AI to analyse fan sentiment

To address the challenge of validating the Premier League’s global appeal, Firemind implemented a solution leveraging generative AI to analyse 12,000 tweets from X (formerly Twitter) that mentioned the Premier League. The project aimed to extract insights into fan sentiment, engagement, and perceptions of the league. The analysis focused on categorising sentiment (positive, neutral, or negative) and measuring engagement metrics such as impressions, likes, retweets, and replies, providing a detailed view of how fans interact with and perceive the league. 

The solution used Amazon Bedrock to host a pre-trained Large Language Model (LLM) for advanced sentiment analysis and topic categorisation. To process the data efficiently, Firemind employed AWS Lambda for serverless compute tasks and Amazon DynamoDB to store tweet attributes and facilitate efficient mapping of IDs to processed data. Data was ingested and stored in Amazon S3, enabling scalability and reliability for handling large datasets. 

A relevance flag was applied to filter out irrelevant or promotional content, such as illegal ticket sales and betting advertisements. This preprocessing step ensured that the AI worked only with high-quality data, enhancing the accuracy of the insights. The filtered tweets were then analysed to detect sentiment trends, tone, and topics, revealing emotional and psychological themes tied to the Premier League’s appeal. 

Visualisations of the results were created using custom charts and graphs, showcasing key trends and insights. Firemind also identified opportunities to enhance future workflows by recommending Amazon QuickSight for seamless integration with AWS-hosted data and streamlined visualisation.

This project demonstrated the Premier League’s ability to connect with fans on an emotional and psychological level. It quantified the league’s global appeal through structured analysis of unstructured data, setting the foundation for scaling the system to analyse 1 million posts per month and integrating other social media platforms, including video and imagery content. 

Comprehensive fan insights

Analysed 12,000 tweets from X to uncover patterns in fan sentiment and engagement. Advanced sentiment analysis provided a deeper understanding of emotional and psychological connections to the Premier League.

Actionable insights for strategy

Used AWS services to transform unstructured social media data into meaningful insights. These findings offered a foundation for tailoring engagement initiatives and aligning fan perceptions with league objectives.

Scalable and reliable processing

Leveraged AWS Lambda, Amazon DynamoDB, and Amazon S3 for efficient and scalable data processing. This architecture ensured the seamless handling of large datasets and laid the groundwork for analysing up to 1 million posts per month.

Model Spotlight

Anthropic is an artificial intelligence research company based in the San Francisco Bay Area. Founded in 2021, the company focuses on developing safe and ethical AI systems, particularly AI assistants capable of open-ended dialogue and a wide range of tasks. 

Anthropic has created notable models like Claude, and explores techniques such as ‘constitutional AI’ to imbue their AI with robust ethical principles. Led by a team of prominent AI researchers, Anthropic is positioning itself as an emerging leader in the field of beneficial AI development, working to ensure AI capabilities advance in alignment with human values.

Claude 3 Haiku

Anthropic Claude 3 Haiku was chosen for this project due to its advanced capabilities in natural language understanding and sentiment analysis. This large language model (LLM) excels at processing and interpreting complex and nuanced text, which is essential for analysing diverse user-generated content from social media and other platforms. 

Claude 3 Haiku’s ability to generate deep insights from vast amounts of data makes it ideal for uncovering subtle sentiments and perceptions about the Premier League and its competitors. Its advanced natural language processing features enable the model to detect and interpret intricate emotional cues and contextual nuances that traditional analytics might overlook. 

Furthermore, the model’s efficiency in handling large datasets and producing actionable insights quickly aligns well with the project’s need for scalable and rapid analysis. By leveraging Claude 3 Haiku, the project was able to gain a comprehensive understanding of global fan sentiments, validate the Premier League’s claim, and provide detailed, data-driven recommendations for enhancing marketing strategies and fan engagement. 

Learn more
Get in touch

Want to learn more?

Seen a specific case study or insight and want to learn more? Or thinking about your next project? Drop us a message!