Background
A rapidly expanding digital commerce brand serving a broad consumer base across multiple product categories adopted Amazon Bedrock foundation models to deliver conversational product search, AI-generated content, and personalised recommendations, boosting engagement and increasing conversions across its online channels.
The Challenge
The customer needed to modernise its digital commerce platform with AI-driven personalisation to improve shopping experiences and boost conversions. Key challenges included:
- Reliance on a static, rules-based recommendation system that struggled to adapt to new shoppers or emerging trends.
- Need for natural language product search and generative product descriptions to improve discovery and engagement.
- Desire for customer service chatbots and personalisation without the complexity of managing large-scale ML infrastructure.
Our Solution
Rambunct implemented Amazon Bedrock to provide a generative AI layer integrated into the customer’s commerce platform, enabling dynamic personalisation at scale.
- Conversational Product Search: Bedrock foundation models enabled customers to describe products in natural language.
- Personalised Recommendations: Combined real-time session data with Bedrock-powered natural language generation for tailored suggestions.
- AI-Generated Content: Automated product descriptions that adapted tone and style to target demographics.
- Customer Support Chatbots: AI assistants embedded in the website and mobile app to reduce reliance on live agents.
The Result
The solution significantly improved engagement, conversion rates, and operational efficiency.
- 25% increase in click-through rates for recommended products.
- 15% higher sales conversions from AI-powered product search and recommendations.
- 70% reduction in content creation time through automated product description generation.
- 60% improvement in customer service response times, reducing human agent workload.
