Skip to main content Link Menu Expand (external link) Document Search Copy Copied
L200 TechWorkshop - Copilot and Agents at Work

Exercise 02: Copilot Studio with built-in connectors to SAP and Outlook

Case study overview

Zava Retail Group is a retail company with both brick-and-mortar stores and a growing e-commerce presence. Zava is focused on modernizing its operations through AI-driven solutions. The company recently launched an internal AI Center of Excellence (COE) to explore how agentic solutions can streamline workflows, improve decision-making, and enhance customer and employee experiences.

Zava wants to automate real-time inventory checks and trigger restocking workflows when stock levels fall below a threshold. They need this integrated with their SAP system and want the agent to notify store managers so they can create purchase orders. Adele Vance, a store manager at Zava, has been selected to pilot the new AI-powered inventory workflow and receive low-stock alerts from the agent.

Description

You’ll build an agent in Copilot Studio that uses a prebuilt connector to automate the customer’s inventory to solve their issue.

Objectives

After you complete this exercise, you’ll be able to:

  1. Identify when a customer’s capabilities need to be achieved through Copilot Studio.
  2. Build an agent in Copilot Studio that uses a prebuilt connector.
  3. Identify when a prebuilt connector in Copilot Studio will solve the customer’s challenge.
  4. Configure features in Copilot Studio to develop a working agent.

Context for SAP connectivity

For this lab, SAP connectivity is established using the SAP Inventory MCP tool. The MCP tool acts as the bridge between Copilot Studio and SAP, enabling the agent to fetch inventory details directly from the ERP system. It supports flexible queries - inventory can be retrieved by product ID or product name - ensuring accurate and timely stock visibility.

Duration

Estimated time: 15-20 minutes

Table of contents