AWS boosts sales pipeline using generative AI solution built on Amazon Bedrock
AWS Sales and Marketing teams transformed their sales workflows with generative AI to automate repetitive tasks and provide personalized insights, improving sales pipeline management.
- Industry
- Tech & Comms
- Location
- United States
- Published
- August 2024
Reported outcomes
+4.9%
quantified impactOther quantified impact
Strategic outcomes
Primary read
Use case focus
Showing 2 of 2
- 1Sales Workflow Automation
- 2Generative AI for CRM
- AWS developed an AI-powered Account Summaries feature integrated with CRM, using Amazon Bedrock and Amazon Q Business to generate personalized account narratives from diverse structured and unstructured data.
- The solution uses multiple models including Amazon Titan and Anthropic Claude on Amazon Bedrock, combining strengths for summary generation.
- Architecture includes asynchronous processing, modular multi-model selection, robust data indexing, retrieval, and hallucination mitigation strategies.
- Security is enforced with row-level access controls and content deletion after delivery.
- Generated over 100,000 GenAI Account Summaries, saving an average of 35 minutes per summary.
- Sellers reported a 4.9% increase in opportunity value and better preparedness for customer engagements.
- Significant positive impact on teams handling multiple accounts and during account transitions.
Architecture
The system uses asynchronous Lambda functions to handle summarization requests, combining multiple LLM models including Amazon Titan and Anthropic Claude on Amazon Bedrock. It indexes and retrieves diverse data sources using vector similarity and cross-encoder models for RAG-enhanced LLM prompting, with rigorous hallucination mitigation strategies.
Sources & evidence1
AI-generated summary. Verify important details with the linked sources before relying on this case.
Explore related AI use cases
Was this useful?