Personalized Customer Service

In their daily jobs, service desk support agents handle a high volume of incoming requests that require fast and accurate resolution. To solve tickets efficiently, agents must consult multiple sources of information, including internal documentation and guidelines, previous tickets, and client-specific data. This fragmented access to critical information slows down ticket resolution and makes it more difficult to efficiently identify relevant information.

The client's goal was to evaluate feasibility of a unified assistant through a Proof-of-Concept (PoC) to increase agent productivity by streamlining access to critical information, enabling faster resolution of service tickets while maintaining consistent processes and compliance with established guidelines.

We leveraged the Azure GenAI stack and our Agent architecture expertise to develop a PoC assistant capable of finding and extracting past ticket and customer-specific information and developed custom data connectors to unify data sourced from internal customer databases, Jira and Confluence. We designed an enhanced chatbot interface to facilitate communication with the assistant and display retrieved information and ticket context in a more structured and accessible way.

By unifying access to relevant information through an AI assistant, this PoC has the potential to:

• Reduce the time spent per ticket resolution, increasing productivity through faster access to relevant knowledge.
• Improve adherence to guidelines and structured processes by enabling service desk agents to handle similar ticket requests in a consistent and standardized manner.
• Support a growing volume of new tickets, data and changing guidelines or processes.
• Reduce onboarding time for new agents by providing immediate access to knowledge and guidance.