Background

RS Components were interested in exploring the possibility of enhancing customer satisfaction by providing personalized product recommendations that cater to their specific interests.

This included developing a system that could accurately evaluate the relevance of different products and assign them relevance scores based on their similarity. The primary goal being, to create a platform that could provide customers with product recommendations that were aligned with their preferences and shopping history.

Objectives

The primary objective of this proof of concept is to show that it is possible to provide better product recommendations aligned with a customer’s interests.
The relevance between products is assessed by assigning relevance scores.
These scores serve as a measure of similarity between each product pair.

Architecture

Architect enterprise platform solutions for a multi-product recommendation engine  that utilises multiple ML models across multiple cloud providers.

Stakeholder Management

Work with internal stakeholders across the management team to agree on project requirements, build the enterprise architecture, delivery plan and technical approach

AI SME

As an SME, responsible for the strategic direction of the AI tools, solutions and governance to dictate the best approach to achieving business outcomes

Security & Privacy Governance

Discussion and ideation of integrating existing architecture and security governance framework in line with company policy

Frontend Development

Manage and support the development team with the creation of the web frontends based on the high-fidelity designs.

Product Recommendation Engine

We were able to achieve the targeted outcome by creating a data ingestion pipeline from multiple data sources and equipping our multiple ML models to understand and recommend products based on customers.

The engine was built to work across multiple products and services within RS.We introduced a complex algorithm that could analyze various factors and generate relevant recommendations for each customer.

Success metrics were key to the PoC being productionised. The accuracy of the recommendations provided and the level of customer satisfaction achieved.

Generative AI Chatbot

In addition to the recommendation engine we produced a Generative AI chatbot to replace their existing customer service website chat.

Not only is the chatbot able to assist customers after hours, it is also able to look across the product landscape and quickly answer more general questions such as, ‘ what should I buy for a home security system on a budget of £50’.