ABN AMRO Bank migrates its Anna and Abby chatbots to Microsoft Copilot Studio, increasing Dutch intent-recognition accuracy by 7%

Overview

ABN AMRO Bank, one of the largest banks in the Netherlands, migrated its customer chatbot Anna and internal employee chatbot Abby to Microsoft Copilot Studio in a six-month project delivered with Microsoft and Capgemini. The new agents use Azure AI Language conversational language understanding (CLU) for intent recognition and entity extraction, with Azure Communication Services handling voice and Azure middleware relaying messages. The Anna agent now handles more than 2 million text conversations and 1.5 million voice conversations with customers every year, covering tasks like unblocking a debit card or changing an ATM withdrawal limit, while Abby supports employees with IT helpdesk and facilities requests. By integrating Azure AI Language CLU with Copilot Studio, ABN AMRO increased its intent-recognition accuracy rate for Dutch by 7%, reduced drop-off and transfer rates, and significantly cut operational and maintenance costs.

The challenge

The infrastructure around ABN AMRO's Anna chatbot faced challenges such as high maintenance and scaling costs, high drop-off and transfer rates, limited capabilities that restricted evolving use cases, and difficulties in natural language understanding (NLU) performance, especially in Dutch. The bank experienced similar challenges with its internal chatbot, Abby.

The solution

ABN AMRO Bank migrated its Anna (customer) and Abby (employee) chatbots to Microsoft Copilot Studio in a six-month project delivered with Microsoft and Capgemini, in a stepped approach: migrating the chatbot infrastructure to Copilot Studio, setting up a dialog manager to recognize intents and extract entities, and implementing continuous integration and delivery. The middleware layer enhances interactions using Copilot Studio as the dialog manager, integrated with Azure AI Language conversational language understanding (CLU) for intent recognition and entity extraction; the text channel passes chat interactions from the bank's contact-center-as-a-service platform to Azure middleware, while voice calls are processed in Azure Communication Services before reaching the middleware. The Anna agent handles more than 2 million text conversations and 1.5 million voice conversations with customers every year, covering tasks like unblocking a debit card or changing an ATM withdrawal limit, while the Abby agent supports employees with IT Helpdesk and facilities requests. The bank uses Azure DevOps with Power Platform Build Tools for continuous integration and delivery, and Power BI for analytics.

Reported business value

By integrating Azure AI Language CLU with Copilot Studio, ABN AMRO increased its accuracy rate in intent recognition by 7% for Dutch, leading to more precise and reliable customer and employee interactions. The transition to Copilot Studio significantly reduced operational and maintenance costs and reduced drop-off and transfer rates, while allowing ABN AMRO to scale its AI capabilities more effectively.