AB-250: Transforming Contact Center Experiences with AI in Dynamics 365

To start the case study.

Background

Fabrikam, Inc. is a mid-sized global company that provides outsourced customer support services for retail and telecommunications clients. The company operates multiple regional contact centers. Fabrikam, Inc. is transitioning to a unified cloud-based platform by implementing Microsoft Dynamics 365 Contact Center.

The goal is to improve customer satisfaction, reduce average handling time, and increase workforce productivity while enabling AI-assisted customer interactions.

Fabrikam, Inc. has an established Microsoft ecosystem, including Microsoft Azure and Microsoft Power Platform. The company requires that all implementations follow best practices for scalability, maintainability, and governance, while aligning with internal governance and security standards.

Current environment

Contact Center infrastructure

• Fabrikam, Inc. currently uses an outdated on-premises telephony system that does NOT integrate with modern CRM systems.

• The company is migrating to Dynamics 365 Contact Center to support voice-based interactions.

• Azure resources are centrally managed by the IT team.

• Azure Application Insights is provisioned but not yet connected to any contact center workload.

• Microsoft Entra ID is used for identity and access management.

Application lifecycle management (ALM)

• The company maintains separate development, test, and production environments in Microsoft Dataverse.

• Solutions are deployed as managed or unmanaged depending on the environment.

• Configuration data must be moved between environments without manual recreation.

• The IT team requires repeatable and governed deployment processes.

Users and roles

• Customer service representatives (CSRs) handle inbound customer calls.

• Supervisors monitor performance and manage workforce scheduling.

• Administrators configure and maintain the system.

• Roles are defined for CSRs, supervisors, and administrators, but permissions are not consistently enforced.

• Access must be restricted based on user responsibilities to enforce least privilege.

Business requirements

Voice channel

• Customers must be able to contact customer support through an inbound voice channel.

• Incoming calls must be routed to available CSRs based on predefined criteria.

• Customer input must be collected through an automated interaction before routing.

• Some inquiries must be resolved without involving a human representative.

• When escalation is required, CSRs must receive relevant information collected earlier in the interaction.

Workforce management (WFM)

• Supervisors must define shifts and assign schedules for CSRs.

• CSRs must receive interactions during their scheduled shifts only.

• Some shifts must allow for times that do NOT accept phone calls.

• WFM capabilities must be enabled within the contact center solution.

AI and automation

• The solution must support AI-driven capabilities to handle frequently asked customer questions.

• The system must support generative AI for natural, conversational interactions.

• When an issue cannot be resolved automatically, escalation paths must be defined.

Routing and work assignment

• Customer interactions must be managed through a structured process to ensure consistent handling.

• Information collected during the voice agent interactions must be available to CSRs.

• CSRs must be assigned to handle interactions based on their expertise.

Analytics and monitoring

• The company must monitor conversation performance and identify and troubleshoot issues affecting customer experience.

• The solution must provide detailed diagnostic information to support troubleshooting across the interaction lifecycle.

• Supervisors must be able to analyze interaction quality and identify trends and bottlenecks.

Technical requirements

Voice channel configuration

• Dynamics 365 Contact Center must be used to support voice-based customer interactions.

• The solution must support inbound calling through configured phone numbers.

• Voice interactions must be routed in a way that follows best practices for scalability and consistent performance.

IVR and voice agents

• IVR flows must support both naturally spoken input and keypad input.

• Voice agents must be configured by using Copilot Studio.• AI agents must be able to retrieve customer data from backend systems when needed.

Security and access

• Access must be restricted based on user responsibilities and follow the principle of least privilege.

• CSRs must access records relevant to only the interactions they handle.

• Supervisors must have visibility into CSR performance and schedules.

• Administrators must have full access to configure and manage the solution.

ALM and deployment

• Production deployments must use managed solutions.

• Configuration data must be moved between environments by using supported tools and without manual recreation.

• Deployment processes should be automated to ensure consistency and reliability.

Issues

• The current system lacks integration between voice interactions and CRM systems.• CSRs receive calls outside of their scheduled shifts.• Fabrikam, Inc. has no consistent process for deploying configurations across environments.• AI capabilities are underused, resulting in increased interaction volumes.• Some customers interact from a noisy environment, which makes it difficult for CSRs and agents to serve them efficiently.• Supervisors lack visibility into detailed conversation diagnostics.• Access permissions are not consistently aligned with user responsibilities, causing access inconsistencies.