Study guide
9.2 Orchestrating Prebuilt AI in Dynamics 365 F&O and M365 Copilot for Sales/Service
Microsoft Certified: Agentic AI Business Solutions Architect Expert AB-100
Track your progress as you read
Finished sections get a check mark.
1 Chapter 1: Requirements Analysis & Grounding Data
1.1 Assessing Agents for Task Automation, Analytics & Decision-Making 1.2 Grounding Data Evaluation: Accuracy, Relevance, Timeliness, Cleanliness & Availability 1.3 Structuring Enterprise Business Data for AI System Interoperability
2 Chapter 2: AI Strategy, Cloud Adoption Framework & CoE
2.1 Cloud Adoption Framework (CAF) for Azure AI Strategy & Roadmap 2.2 Multi-Agent Architecture: Microsoft 365 Copilot, Copilot Studio & Foundry 2.3 Microsoft AI Center of Excellence (CoE) & Solution Governance 2.4 Developing Use Cases for Prebuilt Agents & the Microsoft Agent Catalog 2.5 Solution Rules & Constraints for Copilot Studio, Microsoft Foundry & Foundry Tools
3 Chapter 3: Model Selection, Prompt Engineering & Dynamics AI
3.1 Model Selection Strategy, Custom Models & Small Language Models (SLMs) 3.2 Enterprise Prompt Engineering Guidelines & Prompt Libraries 3.3 Cross-App AI Strategy for Multi-App Dynamics 365 Solutions
4 Chapter 4: ROI, TCO, Build vs Buy & Model Routing
4.1 ROI Criteria & Total Cost of Ownership (TCO) for AI Solutions 4.2 Comprehensive Financial ROI Analysis for Business Processes 4.3 Build vs Buy vs Extend Architectural Decision Framework 4.4 Dynamic Model Router Architecture & Cost-Performance Optimization
5 Chapter 5: Dynamics 365 AI & Contact Center Design
5.1 Business Terms & Copilot Customization in Customer Experience and Service 5.2 Connectors & Data Flow Design for Copilot in Dynamics 365 Sales 5.3 Agent Integration with Dynamics 365 Contact Center Channels
6 Chapter 6: Agent Archetypes & Copilot Studio Core
6.1 Designing Task Agents, Autonomous Agents & Prompt-and-Response Agents 6.2 Copilot Studio Topic Design, Trigger Phrases & Graceful Fallbacks 6.3 Orchestration Selection: Standard NLP vs CLU vs Generative AI Orchestration
7 Chapter 7: Copilot Studio Advanced Actions & Power Platform
7.1 Designing Agent Flows & Prompt Actions in Copilot Studio 7.2 Power Apps Canvas App Integration, Generative Pages & Agent Feeds 7.3 Power Platform Well-Architected Framework for AI Workloads 7.4 Proposing Microsoft Power Platform AI Features & the AI Hub
8 Chapter 8: Agent Extensibility, MCP & Computer Use
8.1 Microsoft Foundry Custom Models & Foundry Tools Selection 8.2 Agent Extensibility via Model Context Protocol (MCP) in Copilot Studio 8.3 Computer Use Automation & Web UI Interactions in Copilot Studio
9 Chapter 9: Advanced Behaviors & Prebuilt App Orchestration
9.1 Agent Behaviors: Multi-Step Reasoning, Voice Mode & M365 Agent Optimization 9.2 Orchestrating Prebuilt AI in Dynamics 365 F&O and M365 Copilot for Sales/Service 9.3 Orchestrating Prebuilt Copilot AI Features in Dynamics 365 Customer Experience & Service 9.4 Proposing Microsoft 365 Agents for Business Scenarios 9.5 Adding Knowledge Sources to In-App Help & Guidance in Dynamics 365 Finance & Supply Chain
10 Chapter 10: Telemetry, Monitoring & Agent Tuning
10.1 Monitoring Architectures: Azure Monitor, Application Insights & Studio Analytics 10.2 Telemetry Data Interpretation, Performance Metrics & Latency Tuning 10.3 User Feedback Analysis, Backlog Prioritization & Continuous Improvement
11 Chapter 11: Testing Strategy & Model Validation
11.1 Agent Testing Frameworks, Quality Metrics & Groundedness Benchmarks 11.2 Custom AI Model Validation, Safety Criteria & Regression Testing 11.3 Copilot-Assisted Test Case Generation & Cross-Dynamics End-to-End Testing 11.4 Validating Effective Copilot Prompt Practices
12 Chapter 12: ALM for Grounding Data & Copilot Studio Agents
12.1 ALM for Grounding Data, Vector Indexes & Knowledge Synchronization 12.2 ALM for Copilot Studio Agents, Connectors & Solution Packaging 12.3 Environment Strategy & Automated Deployment Pipelines for Studio Agents
13 Chapter 13: ALM for Foundry Agents & Dynamics 365 AI
13.1 ALM for Microsoft Foundry Agents Service & Model Registries 13.2 ALM for Custom AI Models, Fine-Tuning Artifacts & Deployment Slots 13.3 ALM for Dynamics 365 AI in Finance, Supply Chain & Customer Engagement
14 Chapter 14: Agent Identity, Security & Access Controls
14.1 Agent Identity Architecture, Microsoft Entra ID & Authentication Flows 14.2 Access Controls, Role-Based Security & Permissions on Grounding Data 14.3 Secure Data Processing & Grounding Data Isolation
15 Chapter 15: Model Security & Threat Mitigation
15.1 Model Security Architecture & Defense Against Vulnerabilities 15.2 Mitigating Prompt Manipulation, Jailbreaks & Indirect Prompt Injections 15.3 Azure AI Content Safety, Prompt Shields & Guardrails
16 Chapter 16: Responsible AI, Governance, Residency & Audit
16.1 Microsoft Responsible AI Standard: Principles, Assessment & Implementation 16.2 Data Residency, Sovereign Boundaries & Cross-Border Compliance 16.3 Comprehensive Audit Trails, Model Decision Logging & Microsoft Purview 16.4 Designing Governance for Agents: Environment Strategy, Data Policies & Agent Inventory
9.2 Orchestrating Prebuilt AI in Dynamics 365 F&O and M365 Copilot for Sales/Service
Key Takeaways Explain This Topic AI
Prebuilt Copilots in Dynamics 365 Finance & Supply Chain Management deliver role-specific generative intelligence—including collections coordination, 3-way invoice reconciliation, inventory visibility queries, and PO change impact analysis—grounded in ERP transactional ledgers.
Extending Dynamics 365 F&O agent experiences with custom external knowledge requires Dataverse Virtual Entities for real-time bi-directional ERP data access without data duplication, complemented by Power Platform connectors and SharePoint SOP grounding.
Microsoft 365 Copilot for Sales and Copilot for Service provide contextual sidecar intelligence inside Outlook and Microsoft Teams, integrating bi-directionally with Dynamics 365 Sales, Salesforce, ServiceNow, and Zendesk.
Enterprise orchestration of M365 Copilot for Sales/Service mandates strict pass-through security honoring source CRM role-based and field-level permissions, tenant-level governance in the M365 Admin Center, and prompt template extensibility configured in Copilot Studio.
Last updated: September 2026
Orchestrating Prebuilt AI in Dynamics 365 F&O and M365 Copilot for Sales/Service
Quick Answer: Orchestrating enterprise prebuilt AI requires harmonizing native domain copilots in Dynamics 365 Finance & Operations (collections, invoice matching, inventory visibility, PO impact analysis) with productivity-embedded copilots (M365 Copilot for Sales and Service in Outlook and Teams). Architects extend these capabilities using Dataverse Virtual Entities and custom Copilot Studio actions while enforcing CRM role-based access control, field-level security, and centralized Microsoft 365 Admin Center governance.
Enterprise business processes rarely exist in isolation. A customer inquiry received in Microsoft Outlook often requires inspecting a sales opportunity in Salesforce or Dynamics 365 Sales, checking inventory balances in Dynamics 365 Supply Chain Management, and reviewing outstanding payment disputes in Dynamics 365 Finance.
Rather than building complex agentic solutions entirely from scratch, enterprise architects leverage Microsoft's prebuilt Copilots —orchestrating and extending them across ERP (Finance & Operations), CRM (Sales & Service), and collaborative productivity surfaces (Outlook & Teams).
1. Orchestrating Prebuilt AI in Dynamics 365 Finance and Supply Chain Management
Dynamics 365 Finance and Supply Chain Management (F&O) embeds prebuilt generative AI capabilities designed to streamline high-volume financial accounting, credit management, procurement, and warehouse operations. +-------------------------------------------------+
- | Dynamics 365 Finance & Operations Platform |
- +-------------------------------------------------+
- |
- +---------------------------+---------------------------+
- | |
- v v
- [ Dynamics 365 Finance ] [ Dynamics 365 Supply Chain ]
- Collections Coordinator Copilot - Inventory Visibility Copilot
- Invoice Reconciliation Copilot - PO Change Impact Analysis
- Financial Reporting Variance Analysis - Demand Forecasting Assistance
- | |
- +---------------------------+---------------------------+
- |
- v
- [ Extensibility & Custom Grounding ]
- Dataverse Virtual Entities (OData)
- External SQL / Fabric Lakehouse
- SharePoint Standard Operating Procedures (SOPs)
- Copilot Studio Custom Actions
- 1.1 Finance AI Capabilities
Financial workflows require absolute auditability and strict transactional precision. Prebuilt Copilots in Dynamics 365 Finance augment human accountants without bypassing accounting rules:
Collections Coordinator Copilot:
Automates credit management and account receivables (AR) outreach.
Ingests customer aging balances, historical payment behavior, credit limits, and dispute logs to generate comprehensive credit risk summaries.
Drafts personalized collection communication emails tailored to the customer's payment profile, proposing structured payment arrangements and attaching relevant open invoices.
Invoice Reconciliation Copilot:
Addresses discrepancies in accounts payable (AP) invoice processing.
Leverages Azure AI Document Intelligence optical character recognition (OCR) paired with generative reasoning to perform automated 3-Way Matching across Purchase Orders (POs), Product Receipts, and Vendor Invoices.
Identifies variance root causes (e.g., unit price discrepancies, unbilled freight charges, partial deliveries) and proposes line-item adjustments or flags invoices for managerial review based on tolerance thresholds.
Financial Reporting Variance Analysis:
Empowers controllers and financial analysts to ask natural language questions regarding financial statements (e.g., "Explain the 18% variance in operating expenses for Q3 across the EMEA cost centers").
Decomposes general ledger (GL) journal transactions, currency fluctuations, and operational anomalies into narrative variance summaries suitable for executive reporting.
1.2 Supply Chain AI Capabilities
Modern supply chains face frequent disruptions in lead times, material availability, and freight transit. Dynamics 365 Supply Chain Management deploys specialized copilots to maintain operational continuity:
Inventory Visibility Copilot:
Provides a high-throughput, microservice-based querying engine over global stock levels.
Allows planners and customer service representatives to query real-time omnichannel on-hand stock, Available-to-Promise (ATP), and allocated reserves using conversational natural language ("Do we have 500 units of SKU-8840 available for immediate shipment in the Midwest distribution network?").
Purchase Order Change Impact Analysis:
When a supplier issues an unexpected delivery delay or quantity reduction on an open PO, the copilot models upstream and downstream ripple effects.
Analyzes which production orders, manufacturing assembly lines, transfer orders, and confirmed customer sales orders are jeopardized by the change.
Recommends mitigation actions, such as rerouting stock from safety inventory, sourcing from alternative approved vendors, or re-sequencing manufacturing job schedules.
Demand Forecasting Assistance:
Augments Azure Machine Learning demand forecasting models by generating natural language insights on seasonal trends, promotional uplift anomalies, and demand spikes, assisting planners in adjusting baseline forecast models.
1.3 Prebuilt Copilots in Dynamics 365 F&O Summary
Prebuilt Copilot Feature ERP Functional Area Primary Business Value Key Grounding Entities
Collections Coordinator Dynamics 365 Finance Accelerates cash collection; reduces Days Sales Outstanding (DSO) CustTable , CustTrans , CustCollectionsCase
Invoice Reconciliation Dynamics 365 Finance Automates 3-way invoice matching; eliminates AP bottlenecks VendInvoiceInfoTable , PurchTable , VendReceiptInfoTable
Variance Analysis Dynamics 365 Finance Produces executive narrative explanations of financial ledger variances GeneralJournalEntry , MainAccount , DimensionAttribute
Inventory Visibility Dynamics 365 Supply Chain Sub-second omnichannel inventory queries and ATP calculation InventOnHand , Inventory Visibility Microservice
PO Change Impact Dynamics 365 Supply Chain Identifies manufacturing and customer order risks from vendor delays PurchLine , ProdTable , SalesLine
2. Extending F&O Agent Chats with Custom External Knowledge Sources
While prebuilt F&O Copilots understand transactional ERP tables, real-world business decisions require contextual data stored outside the core ERP ledger—such as warehouse operational standard operating procedures (SOPs) on SharePoint, quality control telemetry in Azure SQL, or supplier master agreements. [ User in Dynamics 365 F&O In-App Copilot ]
- |
- v
- [ Copilot Studio Extensibility Layer ]
- |
- +-----------+-----------+
- | |
- v v
- [ Dataverse Virtual Entities ] [ External Knowledge Grounding ]
- (Real-time OData to F&O tables) - SharePoint Document Libraries (SOPs)
- CustTable / PurchTable - Azure SQL / Fabric Lakehouse
- No data replication - Third-party REST APIs via Connectors
- 2.1 Dataverse Virtual Entities vs. Dual-Write
To enable Copilot Studio agents to reason over F&O data alongside external knowledge, architects must understand the integration topology:
Dual-Write: An asynchronous/synchronous multi-master replication pattern that copies data between Dynamics 365 F&O and Dataverse tables. While ideal for transactional synchronization between Dynamics 365 Sales and F&O, replicating massive historical tables solely for agent grounding introduces storage bloat and sync latency.
Dataverse Virtual Entities: The preferred architectural pattern for agent grounding . F&O tables are exposed as native Dataverse entities via an OData v4 virtual entity provider.
Zero Data Duplication: Data remains in the F&O transactional store; queries are executed in real time.
Full Copilot Studio Compatibility: Copilot Studio treats virtual entities as native Dataverse tables, allowing generative agents to perform real-time reads and parameterized lookups effortlessly.
2.2 Grounding on Operational SOPs and External Stores
Architects extend in-app F&O copilot experiences by registering custom knowledge plugins in Copilot Studio:
SharePoint Operational SOPs: Ingest warehouse picking guidelines, hazardous material handling manuals, and safety compliance protocols. When a warehouse worker in F&O asks, "How should I handle a damaged lithium battery package during intake?", the copilot cites the authoritative SharePoint SOP alongside ERP return instructions.
External SQL & Fabric Lakehouses: Connect historical maintenance logs or machine sensor telemetry using Power Platform certified connectors, providing holistic context during supply chain planning.
3. Orchestrating Microsoft 365 Copilot for Sales & Copilot for Service
Sales representatives and customer service agents spend the majority of their working day inside communication clients: Microsoft Outlook and Microsoft Teams . Requiring users to constantly toggle between email, chat, and specialized CRM/ITSM web applications disrupts productivity.
Microsoft 365 Copilot for Sales and Microsoft 365 Copilot for Service embed role-specific agentic intelligence directly into these communication hubs, acting as a dynamic sidecar. +--------------------------------------------------------+
- | Productivity Surfaces: Microsoft Outlook & Teams |
- +--------------------------------------------------------+
- |
- +-------------------------+-------------------------+
- | |
- v v
- [ M365 Copilot for Sales ] [ M365 Copilot for Service ]
- Email thread summarization - Inbound case triage & summary
- Opportunity / Contact CRM creation - Solution drafting from Knowledge Bases
- Real-time meeting insights & battlecards - Swarming collaboration in Teams
- | |
- +-------------------------+-------------------------+
- |
- v
- [ Federated CRM & Knowledge Ingestion Layer ]
- CRM: Dynamics 365 Sales OR Salesforce Sales Cloud
- Knowledge: ServiceNow, Zendesk, Dynamics 365 Service, SharePoint
- |
- v
- [ Security & Governance: Pass-Through Permissions & M365 Admin ]
- 3.1 Architectural Scope: Sales vs. Service
- Dimension Microsoft 365 Copilot for Sales Microsoft 365 Copilot for Service
- Target User Account executives, BDRs, sales managers Tier 1-3 support agents, contact center reps
- Host Applications Microsoft Outlook, Microsoft Teams Microsoft Outlook, Microsoft Teams, Web Agent
Supported CRMs Dynamics 365 Sales, Salesforce Sales Cloud Dynamics 365 Customer Service, Salesforce Service Cloud
Supported Knowledge Bases CRM sales collateral, SharePoint ServiceNow, Zendesk, Dynamics 365 KM, SharePoint
Key In-App Actions Summarize email deals, save email to CRM, update opportunity stage, generate meeting prep cards Summarize customer incident email, draft resolution from KB, create case, generate handover notes
Real-Time Meeting AI Live Teams transcription tagging competitor mentions, budget talk, sentiment, and action items Live call transcription, customer sentiment tracking, automated case resolution summary
3.2 CRM & Knowledge Base Grounding Mechanics
Neither Copilot for Sales nor Copilot for Service stores a shadow copy of your CRM database. Instead, they operate as a federated semantic query engine :
User Delegation & Pass-Through Authentication: When a sales rep opens an email from a prospect in Outlook, Copilot connects to the configured CRM (Dynamics 365 Sales or Salesforce) using the rep's OAuth 2.0 delegated Entra ID / CRM credentials .
Security & Permission Inheritance:
Row-Level Security: The agent can only retrieve opportunities, accounts, and contacts owned by or shared with the user.
Field-Level Security: If an organization hides a proprietary field (e.g., Commission_Rate__c or Profit_Margin ) from standard reps via CRM field-level security, the Copilot engine respects this restriction. The field is never included in prompt context or generated summaries.
Federated Knowledge Retrieval in Service: Copilot for Service connects concurrently to external ITSM knowledge bases (e.g., ServiceNow , Zendesk ). When drafting a case resolution, the agent searches both local CRM articles and third-party knowledge bases, presenting synthesized solutions with direct citations to the source knowledge base articles.
4. Administrative Controls, Security & Custom Prompt Templates
Enterprise rollout of prebuilt Copilots requires stringent governance managed through the Microsoft 365 Admin Center and Copilot Studio Extensibility . +---------------------------------------+
- | Microsoft 365 Admin Center |
- | - Tenant-wide Copilot app deployment |
- | - CRM Instance Binding (D365 / SFDC) |
- | - Security & Data Sharing Policies |
- | - Audit Logging & Purview Compliance |
- +---------------------------------------+
- |
- v
- +---------------------------------------+
- | Copilot Studio Extension |
- | - Custom Prompt Templates & Skills |
- | - Certified Custom Connectors |
- | - Power Automate Action Bindings |
- +---------------------------------------+
- 4.1 Centralized Governance in Microsoft 365 Admin Center
Enterprise administrators manage Copilot for Sales and Service through the Microsoft 365 Admin Center under Settings > Copilot :
CRM Instance Binding: Administrators bind the tenant to the authoritative CRM environment (specifying the production Dynamics 365 organization URL or Salesforce production login URL). Mixed or unauthorized developer sandboxes can be blocked.
Tenant-Level Feature Toggles: Control whether AI-generated meeting summaries, automatic email drafting, or external CRM contact creation are enabled for all users or targeted security groups.
Audit Logging and Compliance: Every generative inquiry, CRM record update, and email generation event is captured in the Microsoft Purview Unified Audit Log , recording user identity, timestamp, client IP, and target CRM record ID.
4.2 Role-Tailored Prompt Templates and Extensibility
Organizations often require Copilot to follow company-specific sales methodologies (e.g., MEDDPICC, BANT, Challenger Sale) or service triage protocols.
Customizing Prompt Templates in Copilot Studio: Administrators open the Copilot Studio extension for Sales/Service to modify prompt templates. For example, the default meeting recap prompt can be augmented with: When analyzing this sales call, structure the executive summary using the MEDDPICC framework:
1. Metrics: Quantifiable value expected by customer.
2. Economic Buyer: Identity and role of decision-maker.
3. Decision Criteria: Technical and business requirements.
4. Identify Pain: Core operational bottleneck.
Highlight any missing qualification criteria as an immediate action item.
Extending with Power Platform Connectors: Architects can bind custom actions (e.g., querying credit scores from Experian or checking custom inventory databases) directly into the Copilot for Sales/Service interface, enabling users to execute enterprise workflows without leaving Outlook or Teams.
Important
AB-100 Exam Tip: CRM Permissions Enforcement Copilot for Sales and Copilot for Service never bypass CRM security models. If an exam scenario asks how to restrict sales reps from seeing confidential deal margins in Copilot-generated email summaries, the answer is always to configure native CRM Field-Level Security (FLS) or Dataverse column security. Copilot automatically inherits and honors these permissions via delegated authentication.
Tip
AB-100 Exam Tip: Dataverse Virtual Entities vs Dual-Write for F&O When extending Copilot Studio agents to query large volumes of Dynamics 365 Finance and Supply Chain data in real time without data synchronization overhead, choose Dataverse Virtual Entities . Choose Dual-Write only when ongoing, bidirectional, offline-capable database synchronization is required between Dynamics 365 Customer Engagement and F&O.