Microsoft Enhances Dynamics 365 With FetchXML Agent Routing

Latest Updates

Microsoft released FetchXML-driven customization for Dynamics 365 Contact Center agent transfers and consults, reaching general availability January 26. Administrators can now configure FetchXML queries for four scenarios: consult to queue, consult with a representative, transfer to a queue, and transfer to a representative.

The feature lets administrators manage which queues or representatives are available during consult or transfer actions, part of Release Wave 2 2025 rolling out from October 2025 to March 2026.

Technical Implementation

Building on this declarative approach, FetchXML is the native declarative query language for Dataverse entities used in Dynamics 365. Using FetchXML gives administrators flexibility to target combinations of attributes including custom fields added by partners without requiring code changes to the routing engine.

The FetchXML-driven consult/transfer feature enables targeted handoffs, compliance-aware routing, and smarter consult workflows without code-heavy changes to the routing stack. Organizations can use FetchXML to filter representatives by skills, certifications, or compliance flags to meet regulatory requirements.

The consult/transfer flow continues to follow existing voice-channel settings for placing callers on hold and transfer semantics.

Configuration is managed through the Consult and transfer page in the Copilot Service admin center. Microsoft Message Center note MC1221931 announced the feature on January 23.

The declarative approach means administrators define routing logic through query syntax rather than custom code, reducing implementation complexity and maintenance overhead for contact center operations teams.

Beyond immediate operational benefits, the declarative configuration model positions Microsoft to reduce total cost of ownership for contact center operations. By eliminating custom code deployment cycles when routing rules change, organizations can respond to compliance requirements or skill availability shifts within hours rather than weeks.

For multi-region enterprises managing different regulatory frameworks across jurisdictions, routing logic adapts to local data residency and certification requirements without requiring separate codebases.

AI Agent Capabilities

While query-driven routing improves efficiency, Release Wave 2 includes several AI-driven agents that automate customer service tasks. The enhancement builds on earlier efforts when Microsoft introduced its AI-powered Dynamics 365 Contact Center in June 2024.

The Customer Knowledge Management Agent can generate articles in real-time as agents close cases, responding to emerging issues in minutes, and can process historical cases to rapidly build out knowledge base. The agent analyzes case notes, emails, and conversations to draft articles, check for duplicates, and can be configured to remove sensitive data before publishing.

The Case Management Agent automatically creates cases and populates them with relevant details from the moment a live chat starts or email arrives. It intelligently classifies incoming emails to filter out irrelevant messages, then suggests the right internal expert and initiates a Teams chat with a pre-populated case summary when agents need specialist help.

The combination of real-time knowledge generation and automated case management creates a feedback loop that accelerates organizational learning. As the Knowledge Management Agent captures resolution patterns from closed cases, the Case Management Agent leverages that growing knowledge base to improve initial case classification and expert routing.

These capabilities extend the foundation Microsoft established when it unveiled Copilot for Service at Ignite 2023. This compound effect indicates Microsoft’s strategy to build self-improving support systems where each customer interaction strengthens the platform’s ability to handle future requests efficiently.

Intelligent Routing

Complementing these automation capabilities, intent-based routing uses generative AI to analyze customer messages and understand their true intent, then matches to pre-configured agent groups with specific skills. This approach differs from keyword-based routing by understanding context and nuance in customer requests.

Copilot-powered email template recommendations analyze conversation context to automatically suggest relevant email templates. Agents can enter custom prompts to guide Copilot template suggestions, while administrators can embed pre-defined Copilot prompts directly into email templates that automatically run when agents select the template.

The shift from keyword-based to intent-based routing positions Microsoft to handle the ambiguity inherent in natural language customer requests. Organizations accumulate interaction data that improves classification accuracy and routing precision, reducing misdirected contacts that waste both customer and agent time.

Extensibility

Looking beyond Microsoft’s built-in capabilities, Model Context Protocol (MCP) server enables connection of third-party AI agents like Claude or custom bots built with Copilot Studio to Dynamics 365 Customer Service. External AI agents connected via MCP can perform actions within Customer Service such as creating cases, updating accounts, or drafting emails.

This extensibility supports hybrid AI strategies where organizations combine Microsoft’s built-in agents with custom or third-party solutions tailored to industry-specific workflows.

Workspace Improvements

Meanwhile, enhancements to the agent workspace address operational continuity challenges. Customer Service workspace automatically restores all previously opened sessions and tabs after browser refresh or reconnection.

Multisession capabilities from Customer Service Workspace can now be enabled for any custom model-driven app, extending the productivity benefits of parallel session management beyond the standard workspace environment.

Supervisors can filter active conversations in Omnichannel real-time analytics dashboard based on attributes like queue, status, or customer intent, then perform bulk actions on conversations. Supervisors can reassign multiple conversations to another queue, send notifications to assigned agents, or bulk-close inactive threads.

Administrators can add line-of-business (LOB) tags to email templates in bulk and create custom views.

The workspace enhancements collectively address the fragmented nature of contact center operations. Session persistence eliminates the productivity loss from network disruptions or browser crashes, while bulk conversation management gives supervisors the leverage needed to manage hundreds of concurrent interactions.

For enterprise contact centers handling peak volumes during product launches or service disruptions, the ability to reassign dozens of conversations with a single action far exceeds the manual queue management capabilities of legacy systems.

Industry Context

Despite these platform advancements, industry analysts temper expectations for immediate transformation. Microsoft was named a Leader in the 2025 Gartner Magic Quadrant for CRM Customer Engagement Center, positioning the company to capitalize on growing enterprise demand for AI-powered service platforms.

However, Forrester predicts 2026 will be defined by foundational work for AI in customer service rather than dazzling transformation.

“Customer service leaders are counting on an impending year of business transformation boosted by the power of today’s AI. Yet 2026 will not be that year. Instead of dazzling transformation, the year ahead will be defined by gritty, foundational work — the kind that rarely makes headlines but is critical to realizing AI’s long-term promise.”

Forrester Research (via Forrester Research)

Microsoft’s incremental feature releases align with this pragmatic industry trajectory. The analyst firm predicts one in four brands will see a 10% increase in successful simple self-service interactions by the end of 2026, while daily agent workloads drop by an average of 1 hour as AI automates narrow tasks such as creating FAQs.

Forrester also projects that 30% of enterprises will create parallel AI functions that mirror human service roles including managers to onboard and coach AI agents. This organizational evolution reflects the need for dedicated teams to configure, monitor, and optimize AI agent performance as these systems move from experimental to operational status.

“2026 won’t be the year that AI transforms customer service operations. Instead, it will be the year of hard work — of simplifying, restructuring, and preparing. The organizations that embrace this reality will be best positioned to deliver the AI-first experiences that customers increasingly expect.”

Forrester Research, Predictions 2026: AI Gets Real For Customer Service (via Forrester Research)

This measured outlook gains support from Forrester’s State of AI Survey 2025, which shows that 78% of AI decision-makers find AI outputs trustworthy, suggesting the technology readiness exists even as organizational readiness lags. The gap between technical capability and operational implementation explains why 2026 will focus on infrastructure rather than transformation.

Administrative Features

Reflecting this emphasis on operational readiness, data sensitivity labels allow service representatives to apply Microsoft Purview labels directly to outgoing emails. This integration with Microsoft’s data governance platform helps organizations maintain compliance with data protection regulations by classifying customer communications based on sensitivity levels.

Complementing these governance capabilities, enhanced multi-language support for email templates is included in the November 2025 update. Additionally, Copilot embedded summaries – which automatically generate conversation summaries for agents — are on by default for all copilot-enabled organizations at general availability starting from September 2025.

Unified Service Desk will be deprecated beginning April 01, 2026. Microsoft recommends that users start using Copilot Service workspace as replacement for Unified Service Desk.

Organizations still running Unified Service Desk have less than three months remaining to plan and execute migration to the modern workspace environment.