Targeting and Personalisation: Predictive Insights Drive Higher Transaction Value for eCommerce Platform
Industry: eCommerce
Challenge: Developing a Predictive Next Best Offer Solution
Solution: Machine learning and predictive analytics
Technology: AWS, Redshift, Python, MongoDB
Situation and Challenge
A live entertainment and event ticketing technology company, managing millions of customer interactions across multiple channels, sought to:
- Improve channel integration for enhanced customer experience
- Understand customer preferences and buying habits across channels
- Create targeted and personalised offers to increase average transaction value
- Enhance up-selling and cross-selling of complementary products
- Boost customer satisfaction and loyalty
Insight and Action
Our approach encompassed:
- Evaluating various algorithms to determine the best fit for the complex, high-volume dataset
- Selecting the Naïve Bayes algorithm for its speed, simplicity, and effectiveness with large, feature-rich datasets
- Implementing a multi-step analysis process:
– Feature Selection
– Model Training and Evaluation
– Model Deployment
– Personalised Offer Generation
- Deploying and managing multiple predictive models, generating weekly scores for millions of customer interactions
Our intervention delivered substantial benefits:
– Achieved a 15% increase in conversion rates through personalised next best offer solutions
– Realised an immediate 13% increase in average order value following model deployment
– Enabled real-time predictions for individual customers, facilitating effective up-selling and cross-selling
– Provided insights into customer behaviour and preferences, informing future marketing and sales strategies
Key Achievements:
– Developed and deployed dozens of predictive models
– Created a scalable solution handling millions of customer interactions
– Implemented real-time personalisation capabilities
– Enhanced long-term customer satisfaction and purchase likelihood
Call us On +44 (0) 208 004 3015