Talk to Your ERP: How the Model Context Protocol (MCP) Is Redefining Supply Chain System Interaction

In This Article

Since the beginning of software, users have desired to ask a question of their systems and for that system to give them the answer. Instead, for decades, we have been forced into learning to navigate the confusing menu items of multiple systems and calculate the answer in Excel to ascertain the answer to a question.

Now, your ERP can respond on its own!

Imagine saying, “Hey Copilot, what procurement features are turned on in the test environment, that is turned off in production?” and watching it happen… no IT tickets, no twelve-tab detours, no guessing which menu hides the setting.

That’s the promise of Microsoft’s Model Context Protocol (MCP). It’s not another AI buzzword; it’s the quiet architecture that lets Copilot (large language models) do the clicking and legwork for you, so you can focus on more important work. For manufacturers and distributors, that shift isn’t just convenient. It’s transformative. It means faster troubleshooting, smarter configuration, and one less reason to fear your ERP.

What is the Model Context Protocol?

The Model Context Protocol, or MCP, is a framework that enhances the integration of AI applications and services. For manufacturers and distributors, this translates to faster decision-making, more accurate system changes, and safer adoption of AI at scale.

Essentially, it makes it easier for AI tools to retrieve real-time data, perform actions, and work more reliably across different environments. This means faster, easier, and safer adoption of cutting-edge AI.

That’s why top innovators, like Microsoft, are investing heavily. Recent key capabilities with the MCP include:

And, perhaps most importantly for manufacturers and distributors, Microsoft offers some compelling MCP-enabled use cases in Dynamics 365 Finance & Supply Chain Management (F&SCM), specifically.

The MCP Inside Dynamics 365 Finance & Supply Chain

Within Dynamics 365 Finance & Supply Chain Management, MCP allows teams to engage with their ERP in increasingly interactive, conversational, and streamlined ways.

For example, AI Agents can now check whether environments are configured consistently, compare setups between sites, or provide intelligent recommendations for improvement. This helps supply chain organizations scale more efficiently across legal entities, plants, or regions.

Agents can also execute hands-on system actions like releasing recent records, verifying feature settings, submitting items to workflow, or updating customer and supplier records. Imagine saying, “Disable that inactive supplier,” and having the system do it on your behalf.

They can even answer operational questions like, “What’s the current lead time for X component?” or “Which orders missed their ship date this week?”

Essentially, with the MCP, you can talk to your ERP to accomplish your work.

Budgeting Around MCP-Enabled Tools

While the opportunities are certainly exciting, it’s important to remember that AI is, in fact, an investment. First, you’ll need to understand the benefits and tradeoffs related to using an AI-only approach vs a hybrid approach: MCP, application programming interface (API), or a blended approach.

Under the hood, the MCP server for Dynamics 365 allows the agent to perform actions on your behalf by mimicking the clicks and steps a user would normally take. It’s flexible and fast to adopt as you don’t need to build additional integrations, however run costs may be higher than a traditional approach.

On the other hand, Direct API connections, the traditional approach, route the agent to the system’s underlying interfaces directly instead of issuing multiple MCP calls to mimic every user interaction. This approach takes more effort up front, but once in place, it’s typically more efficient. This means it’s often a better choice for high-frequency or repeatable tasks, where the MCP could become cost-prohibitive.

An MCP approach might be more beneficial for occasional workloads and a traditional API approach for high-volume repeated transactions with little variability. Put another way, MCP is like Uber where the Uber driver does the actions for you at a higher cost than owning, maintaining, and driving the car yourself.

In most manufacturing and distribution contexts, a hybrid model tends to work well. Teams can rely on APIs for processes that run often or at scale, then leverage MCP for occasional or exploratory use cases.

Ultimately, the right choice will depend on frequency, complexity, and long-term needs.

Implementing the MCP

If you’ve decided you’d like to explore the MCP (especially within Dynamics 365), it’s essential to choose a partner with real manufacturing and supply chain expertise in addition to AI expertise.

MCA Connect is already leveraging Microsoft’s new Agent Foundation, combining MCP capabilities with Copilot Studio.

Instead of relying on manual effort or basic scripts, MCA Connect applies AI-driven automation to automate, accelerate, and de-risk the process of implementing, maintaining, and using Dynamics 365.

With automated detection of configuration mismatches, teams can resolve issues before they impact the business, dramatically reducing troubleshooting effort and the familiar “it worked in test, but broke in prod” headaches. And because this approach is both scalable and repeatable, it works reliably across clients, projects, and environment footprints, even as organizations grow or add new solutions.

Leveraging the MCP in Manufacturing and Distribution

We’ve come a long way from command lines and keystrokes. Now, a planner can literally talk to the system that runs their supply chain—and it can provide answers.

That’s the real shift behind the Model Context Protocol. It’s not just a new integration layer; it’s the bridge between human intent and system execution. For manufacturers and distributors, that means less time wrestling with configurations and more time optimizing production, fulfillment, and customer commitments.

The next phase of operational excellence won’t come from adding more screens—it’ll come ease of use through easier interaction. When your ERP becomes a conversation instead of a chore, efficiency stops being a project and starts being the way you work.

Ready to see how conversational AI can move your supply chain forward? Schedule a complimentary AI Use Case Envisioning Session with me and uncover where MCP-enabled efficiency starts delivering results.