How Retail AI helped RBFA, build modern Recommender Systems and increase customer satisfaction

At the recent Devoteam G Cloud Day 2023, experts delved into the fascinating world of retail AI and its impact on building modern recommender systems. While the focus was on retail, it became evident that the applications of retail AI extend far beyond this industry. This blog post will explore the key insights shared by the RBFA, and uncover how recommender systems can benefit businesses across various sectors.

What is Retail AI and Why Does Your Business Need it?

Let’s start by understanding the concept of retail AI and its relevance across industries. Retail AI refers to the AI technologies application in the retail sector and beyond. It encompasses various AI-powered solutions, including but not limited to recommender systems, chatbots, virtual assistants, and inventory management. Incorporating AI into your business is essential to thrive in today’s competitive landscape. It enables personalised customer experiences, improves operational efficiency, optimises inventory management, and drives data-driven decision-making.

To understand the significance of AI in retail, it’s crucial to examine the core components of retailing: recommendations and search. Recommender systems play a vital role in personalising the shopping experience. They analyse user data, including browsing history, purchase behaviour, and demographic information, to provide tailored recommendations. Search algorithms ensure relevant and accurate results when customers search for products or services. These AI-driven capabilities go beyond traditional retail businesses and find applications in content platforms, streaming services, and even professional industries.

The Importance of Good Recommender Systems

Did you know that businesses lose over $300 billion due to bad online experiences in the US alone? Effective recommender systems can make or break a business. By presenting users with personalised recommendations, businesses can increase customer satisfaction, and engagement, and ultimately drive sales. Studies have shown that a well-designed recommendation system can boost conversion rates, average order value, and customer loyalty. Conversely, poor search experiences and irrelevant recommendations can lead to frustration, decreased customer trust, and financial losses. We’ll dive deeper into the significance of implementing robust recommender systems by presenting compelling numbers and statistics.

Recommender systems find applications far beyond the retail sector. Their versatility is evident in various industries and contexts. For instance, content recommendation algorithms power news aggregators, suggesting relevant articles to readers based on their interests. In professional settings, AI-driven recommender systems assist experts by suggesting relevant technical documents or research papers. Educational platforms leverage recommendation algorithms to provide personalised course suggestions to students. The possibilities are vast, and businesses across industries can leverage recommender systems to enhance user experiences and drive engagement.

Retail AI in Practice: The Digital Strategy of the RBFA

The RBFA embarked on a reinvention journey in 2019, aiming to adopt a more contemporary approach as a modern football organisation. They identified three key objectives: 

• Engaging fans,
• Promoting football development,
• Generating revenue to reinvest in the previous activities.

To accomplish these goals, the RBFA realised  the pivotal role of data, digital platforms, and content in its strategy.

Entertainment and Personalisation

The RBFA application serves as a central hub for engaging with stakeholders. They draw inspiration from platforms like Tomorrowland and Netflix to create a personalised user experience. Just as users expect tailored content on these platforms, the RBFA aims to provide a starting page customised to each user’s interests. 

The Technology Stack

The RBFA relies on a robust technology stack to support its data-driven approach. They have implemented a data lake using Google BigQuery to store and analyse their vast data assets. Firestore, a transactional database, seamlessly integrates with the RBFA application, facilitating real-time interactions. The organisation leverages an omnichannel communication tool to personalise content using its data, employing dynamic content blocks and push notifications. To optimise its advertising, the RBFA utilises Google Ads Manager and a recommender engine, ensuring targeted advertisements reach the right audience.

There were challenges in identifying ticket buyers who were also amateur football players. However, by implementing a single sign-on solution with Okta, the RBFA achieved a breakthrough. The single sign-on allows the RBFA to connect various aspects of an individual’s football life, enabling tailored services based on their interests and roles.

MVP to Continuous Improvement

The RBFA followed a phased approach to application development. They released a minimum viable product (MVP) in 2021, ensuring relevance and timely delivery to coincide with the European Championship, EURO 2020. By continuously enhancing the application, the RBFA adapts to evolving user needs and avoids the delay associated with waiting for a complete product. This iterative approach enables the RBFA to deliver a valuable experience to users while maintaining their commitment to grassroots football.

Enhancing Services and Monetisation

The RBFA emphasises the importance of  balancing grassroots (amateur football) and national team activities. By nurturing engagement at multiple levels, the RBFA maximises the potential of the application. They understand the significant impact of amateur football players as drivers of app usage, showcasing the value of catering to diverse football interests.

Data plays a crucial role in the RBFA’s operations, offering five key benefits: 

• Improved services and products, 
• Better understanding of the stakeholders,
• Informed decision-making, 
• Enhanced business processes, 
• Monetization opportunities. 

The RBFA leverages data to recommend personalised content, products, and advertisements. By obtaining user consent and ensuring GDPR compliance, the RBFA builds trust and creates an environment where users willingly share their data.

Implementation Challenges and Ethical Considerations

While the potential of retail AI and recommender systems is immense, challenges and ethical considerations associated with their implementation must be acknowledged. Issues like data privacy, algorithmic bias, and explainability demand attention to ensure AI systems operate transparently, fairly, and by legal and ethical frameworks. We’ll explore these challenges, discuss potential solutions, and emphasise the importance of responsible AI development.

Conclusion

In conclusion, AI-powered recommender systems have revolutionised the retail industry and extended their reach to various sectors. By personalising user experiences, businesses can unlock significant benefits such as increased customer satisfaction, engagement, and revenue. The case study of the RBFA exemplifies the tangible impact of retail AI in the real world. However, it is essential to address implementation challenges and ethical considerations to build trustworthy AI systems. As we move forward, embracing the potential of AI in retail will be vital for businesses aiming to thrive in the digital age. By harnessing the power of AI, we can create a future where personalised experiences and data-driven decision-making become the new norm.