Role-Based Programme
RB0571

Microsoft Copilot Studio for Finance & Accounting

Build AI Agents for Finance Queries, Approvals & Accounting Workflows

IMAGE REQUIRED
Duration
8 Hours
Level
Basic
Delivery
Instructor-Led
Format
Workshop

Programme Objectives

  • Build foundational proficiency in using Microsoft Copilot Studio for finance and accounting information and workflow scenarios.
  • Create agents grounded in approved finance policies, accounting procedures, expense guidance, payment processes, and organizational knowledge.
  • Apply topics, generative orchestration, tools, connectors, and agent flows to automate repeatable finance interactions.
  • Design practical workflows for invoice queries, expense guidance, approval routing, close-process support, reporting requests, and exception escalation.
  • Apply appropriate controls for financial data, segregation of duties, approvals, accuracy, access, and human validation.

Tools covered

Microsoft Copilot StudioGenerative OrchestrationKnowledge SourcesSharePointDataversePower Platform ConnectorsTopicsToolsAgent FlowsApprovalsEvaluationAnalyticsPower Platform Environments

Who should attend

  • Finance Managers
  • Accounting Managers
  • Financial Controllers
  • Finance Operations Professionals
  • Accounts Payable Professionals
  • Accounts Receivable Professionals
  • General Ledger Professionals
  • Financial Planning & Analysis Professionals
  • Management Accounting Professionals
  • Expense Management Professionals
  • Finance Shared Services Professionals
  • Finance Transformation Professionals

Prerequisites & Participant Readiness

  • Basic understanding of finance and accounting processes
  • Familiarity with invoices, expenses, approvals, reconciliations, or financial reporting
  • Basic awareness of organizational finance policies
  • Familiarity with Microsoft 365 or Power Platform is helpful
  • No coding expertise required
  • Access to Microsoft Copilot Studio is recommended for hands-on activities

TOC Modules

Concepts
  • Understanding agents, instructions, knowledge, topics, tools, and agent flows
  • Identifying finance scenarios suitable for conversational and workflow assistance
  • Understanding the difference between information support, process automation, and financial judgement
  • Recognising activities requiring authorized finance approval
Practical activities
  • Exploring a sample finance-support agent
  • Defining agent purpose, users, and boundaries
  • Mapping Finance Need → Agent Capability → Finance Outcome
  • Creating a basic finance-agent use-case map

Scenarios

Invoice & Payment Status Assistant

User Query → Copilot Studio → Transaction Details → Authorized Finance Data Source → Invoice / Payment Status → Exception Identification → Finance Escalation

Participants design an agent that supports routine invoice and payment-status enquiries while ensuring sensitive transaction information is accessed only through authorized controls.

Employee Expense Policy & Approval Assistant

Employee Request → Approved Expense Policy → Eligibility Guidance → Required Information → Agent Flow → Manager / Finance Approval → Status Update

Participants build a controlled workflow that helps employees understand expense requirements and routes valid requests through appropriate approval stages.

## Current Capability Reference

Microsoft Copilot Studio's current **generative orchestration** uses an LLM-driven planning layer that interprets user intent, breaks complex requests into steps, and selects the appropriate knowledge, topics, and tools. Microsoft notes that the runtime can require approval before sensitive actions are performed, which is especially relevant for financial workflows.

Copilot Studio currently supports enterprise knowledge through sources such as **SharePoint and Dataverse**, while Power Platform connector-based knowledge can also connect to supported enterprise platforms including **Oracle Database, SAP OData, Snowflake, Databricks, Google Sheets, Dynamics 365, and ServiceNow**.

For quality control, Copilot Studio provides structured **agent evaluation** that can test response quality, expected meaning, keywords, and expected tool usage. Evaluation results can also expose transcripts, activity maps, and the resources an agent used, supporting controlled testing before finance-facing deployment.

Current Copilot Studio analytics include **agent effectiveness, usage, custom business metrics, knowledge/tool performance, and estimated time or cost savings**, helping finance teams identify where agent workflows are effective and where further refinement is required.

At enterprise level, Copilot Studio agents are created and managed inside **Power Platform environments**, which define data boundaries, security roles, permitted connectors, data policies, and lifecycle separation. Microsoft also supports controls such as DLP policies, customer-managed encryption keys, environment routing, and restrictions on knowledge sources or agent capabilities.

**Finance & Accounting Principle:** Copilot Studio should support **policy guidance, transaction-status queries, workflow routing, reminders, information retrieval, exception identification, and reporting assistance**. Journal postings, payment authorization, accounting treatment, forecasts, financial statements, credit decisions, tax positions, and final financial approvals should remain under appropriately authorized human control.

Related programmes

Continue with programmes from the same capability area.

Take the next step

Ready to make this programme work for your team?

Customise modules, duration and business scenarios for your team.

Instructor-ledVirtualHybrid

Designed around your roles, tools and real workflows.