MoneySuperMarket deployed Money Concierge AI assistant with Amazon Bedrock

Use case typeCustomer personalizationUpdated Feb 6, 2026

MoneySuperMarket (MONY Group plc) deployed the Money Concierge AI assistant to help customers evaluate credit card offers, eligibility, fees, and interest-free periods. The solution uses Amazon Bedrock, Bedrock Knowledge Bases, Amazon OpenSearch Serverless, Amazon S3, AWS Step Functions, and AWS Lambda in a private, serverless architecture integrated into the credit-card journey.

Organization
MoneySuperMarket
Industry
Finance
Location
United Kingdom
Published
February 2026

Reported outcomes

+93%

positive feedbackCustomer experience

55,000 customerscustomers helped

Strategic outcomes

Customer experience & trustImproved decision support for credit card selectionSpeed & agilityMoved from proof of concept to production in 6 monthsCustomer experience & trustDelivered personalized card suggestions and repayment guidance

Catalog median for customer experience deployments: +69% across 99 reported metrics. Compare benchmarks →

Primary read

Use case focus

Showing 2 of 2

  • 1Customer personalization
  • 2Decision support
Customers found it overwhelming to compare credit card offers and understand eligibility, intro periods, fees, and repayment choices.
  • MoneySuperMarket built an AI-powered assistant on AWS that extracts customer context from forms and credit reports, enriches prompts through Step Functions, grounds responses with Bedrock Knowledge Bases, and uses current offers from third-party APIs.
  • The solution was moved from proof of concept to production in six months and integrated into the credit card channel.
Technologies
Amazon BedrockAmazon Bedrock Knowledge BasesAmazon OpenSearch ServerlessAmazon S3AWS Step FunctionsAWS Lambda
  • The assistant helped over 55,000 customers and received 93% positive feedback.
  • The article reports improved customer engagement and conversion.
Architecture

A serverless AWS architecture used Amazon S3 for documentation and terms, AWS Lambda as the API entry point and VPC-attached compute, AWS Step Functions to orchestrate backend workflows and prompt enrichment, Amazon Bedrock for generative responses, Bedrock Knowledge Bases for retrieval-augmented generation over MSM documentation and T&Cs, and Amazon OpenSearch Serverless as the vector store. VPC endpoints kept Bedrock traffic private within AWS.

Implementation partners1
Firemind
Sources & evidence1
Groundedness: 5/5Type: Blog PostPublished: Feb 6, 2026Publisher: AWSEvidence: VendorConfidence: Medium
Primary source

AI-generated summary. Verify important details with the linked sources before relying on this case.

Explore related AI use cases
Customer evidence
More from MoneySuperMarket
Provider evidence
AWS provider pageAWS ranked cases
Market context
Finance AI use casesFinance industry insightsUnited Kingdom country insights
AI patterns
RAG
Browse the catalog
All AI use casesAI use-case typesCustomer personalization use cases

Was this useful?