This is our story of transforming the customer outreach strategy of a large retail chain by harnessing advanced data analytics and a tailored hybrid recommender system. Through sophisticated customer segmentation and Generative AI (Gen AI)-led marketing, we enhanced operational efficiency, created effective communication and enriched customer experiences.
As we know…
Retailers need a granular understanding of their consumer base to deploy targeted promotions, personalized offers and tailored communications / product recommendations. Without accurate insights, campaigns may miss the mark, leading to wasted resources and diminished customer engagement.
The rise of digitalization and advanced analytics has opened new avenues for retailers to better understand customer preferences and drive sales through tailored recommendations and loyalty programs. In this context, leveraging data-driven insights is critical for maintaining a competitive edge and fostering customer loyalty.
2024
WNS Secures Silver at the Stevie Awards for AI / ML Solutions in Generative AI-enabled Marketing Campaigns
The challenge for our client was…
Despite experiencing overall revenue growth, it faced a multi-faceted challenge in driving wine sales:
Increasing Wine Exploration
Problem Statement:
Customers favored only a few popular wine varieties, overshadowing the broader range of options available. The complexity of wine choices often deterred customers from exploring new varieties, leading to a stagnant sales mix.
Creating Targeted Marketing Campaigns
Problem Statement:
The client's marketing efforts were too broad; it needed to address the distinct preferences of different customer segments. More targeted communication was required to fully grab opportunities to drive diverse wine purchases and enhance customer engagement.
Identifying Customer Segments
Problem Statement:
The client needed a clearer understanding of customer segments for effective engagement and retention, particularly in the pivotal wine category. Strengthening customer loyalty was crucial for improving the Customer Lifetime Value (CLTV) and ensure sustainable business growth.
Stepping in as a retail analytics partner…
WNS Analytics (WNS’ data, analytics and AI practice) adopted a tailored approach to develop an award-winning solution. Our extensive data collaboration with the client enabled efficient collection, structuring and management of a rich dataset essential for personalizing customer experiences.
Data Sources
The dataset comprised:
- Transaction data linked to customer IDs
- Detailed customer attributes
- Comprehensive product information, including descriptions, pricing and ingredients
Data Integration
- Data from various sources was aggregated and integrated into an AWS Redshift SQL environment.
- This robust infrastructure harmonized diverse data sources, creating a reliable foundation for advanced analytics.
Data Security and Privacy
The loyalty program solution only used data that customers voluntarily disclosed and consented to share.
This approach:
- Prioritized privacy
- Enhanced personalization in recommendations
- Enriched customer experience
Subsequently, we conducted comprehensive discussions with the client to understand their wine business. This helped us conceptualize an ideal solution that would mimic the personalized service of an expert wine seller who understands each customer's history and preferences. Other key aspects of the solution approach included:
Solution Approach
Methodology Exploration
- Conducted a comprehensive exploration of various methodologies to develop a robust solution
- Selected the Hybrid Recommender System for its ability to integrate multiple models
Solution Customization
- Recognized sophisticated customer segmentation as a crucial element of the solution to accurately identify distinct groups
Iterative Development
- Used extensive datasets, focusing on recent transactions
- Initially applied the solution to 500,000 customers and 16 million transactions
- Built models on AWS cloud infrastructure for adaptability and scalability
WNS Analytics co-created a sophisticated, data-driven wine recommender system to...
Address the client’s challenges. The solution comprised key components, including:
Customer Segmentation
Utilizing advanced analytics, we segmented customers based on their purchasing patterns, which enabled recommendations tailored to individual tastes and preferences.
Hybrid Recommender System
Our hybrid model integrated various methodologies, identifying purchase patterns and matching wines to preferences.
Gen AI-driven E-mail Personalization
To engage customers beyond in-store interactions, we deployed a Gen AI-driven e-mail personalization strategy that enhanced marketing campaign effectiveness and encouraged broader wine exploration.
AWS SageMaker
AWS Glue
AWS Redshift
AWS S3
Open AI's GPT-3.5 Turbo
The implementation of the recommender system…
Consistently met customer needs and exceeded expectations with personalized interactions. Effectively engaging customers and encouraging exploration of diverse product varieties have long-term benefits in brand loyalty, market expansion and revenue growth.
Key measurable outcomes included:
Enhanced Customer Engagement
~ customers reached via personalized communication
4x increase
in e-mail campaign click-through rates
Operational Efficiency
~ percent reduction in turnaround time for preparing marketing content, enabling dynamic and responsive campaign management
Cost Reduction
percent reduction in content curation costs highlighted the cost efficiency of the automated system over traditional methods