What Is MCP, and Why Are 91% of Manufacturing Leaders Prioritizing It Right Now?

Every major manufacturer has made an AI investment in the last two years. Many have made several. 

Yet product decisions still stall. Engineers still chase specs across five different systems. Change orders still take weeks. 

The tools are there. The intelligence isn't delivering.

The cause of that gap is not mysterious. It’s fragmented data. And a growing number of manufacturing leaders believe MCP is the solution.

New research from Propel Software, based on a survey of 400 senior manufacturing professionals, confirms that MCP has crossed from early adoption into mainstream strategic priority, and that the manufacturers who act now will have a structural advantage over those who don't.

What Is MCP? A Quick Overview for Manufacturing Leaders

Model Context Protocol (MCP) is an open standard that lets AI agents securely connect to, query, and act across enterprise systems without requiring a custom integration for every new connection.

Think of it as a universal adapter for enterprise AI. Before MCP, connecting an AI tool to your PLM required a bespoke integration. Then another one for ERP. Another for CRM. The result was a fragmented AI architecture built on one-off connections, each requiring maintenance, each creating risk.

MCP replaces that model with a single, standardized protocol. When your PLM, ERP, and quality systems all speak MCP, any MCP-compatible AI client — Claude, ChatGPT, or your enterprise AI platform of choice — can access all of them through one governed interface.

For manufacturers, that means a conversational request can pull BOM data, check procurement agreements, qualify alternate parts, and generate a change order across every system involved. All in minutes, not days.

What Did the Research Find? Key MCP Statistics for Manufacturing

Propel commissioned Talker Research to survey 400 senior manufacturing professionals across high tech/electronics, industrial equipment, and medical devices. Respondents ranged from directors and managers to CxOs and board members. Here is what the data shows:

On adoption momentum:

• 91% of manufacturing professionals report increased organizational interest in MCP
• 63% describe that interest as dramatic or significant
• 96% of CxOs and 85% of board members report increased interest — confirming this is a C-suite directive, not an IT experiment

On strategic importance:

• 84% rate MCP as critical or very important to their business strategy
• 85% believe MCP will be transformational or have significant impact within three years
• Over 90% have implemented or expect to implement MCP within 12 months

On the role of connected data:

• 89% agree that a connected product data foundation is essential to realizing AI's full potential
• 75% say it is critical or very important for AI to understand relationships across product, quality, supplier, manufacturing, and customer information
• 66% say AI would be completely or mostly ineffective in their organization without access to product data like requirements, bills of materials, and change history

On the cost of inaction:

• 56% say delayed MCP adoption will result in higher operating costs or competitive disadvantage
• EVPs and VPs estimate that nearly one-third of their daily business processes will be automated or influenced by MCP within three years

Why Is MCP a Priority for Manufacturers Specifically?

Manufacturing is a data-intensive industry built on interconnected decisions. Whether a product ships on time, at cost, and to spec depends on the quality of information flowing between engineering, procurement, quality, operations, and commercial teams at every stage of the product lifecycle.

Most manufacturers have digitized those functions. They have PLM. They have ERP. They have QMS. What they often lack is a way for those systems to communicate with each other in real time, and a way for AI to act on the combined picture.

MCP solves the connectivity problem. But the research points to a more specific finding: connectivity without the right data foundation still fails.

89% of respondents agree that connected product data is essential to realizing AI's full potential. And 66% say AI would be mostly or completely ineffective without access to product-related data specifically. 

That finding points directly at PLM as the system MCP must touch first, a conclusion the research confirms explicitly: PLM ranked first as the top system for AI agents to access through MCP (49%) and first as the most valuable data source for AI actions (38%).

The implication is clear. MCP's value in manufacturing is proportional to the quality and structure of the product data it connects to. Fragmented, siloed, or outdated product records don't become more useful when AI can reach them faster. They become a faster path to bad answers.

What Is the Business Case for MCP Adoption?

The business case for MCP in manufacturing rests on three compounding dynamics.

First, AI effectiveness is being held back by data access, not model capability. The models are capable. The bottleneck is context. When an AI agent can't cross system boundaries to access product, quality, and supplier data simultaneously, it can't reason across the full picture. MCP removes that constraint.

Second, the window for deliberate adoption is closing. More than 90% of manufacturing leaders expect MCP implementation within 12 months. The manufacturers who establish a governed, structured MCP foundation now will have a meaningful head start on process automation, decision speed, and cross-functional data access. Those who wait will be implementing MCP reactively, under pressure, with less time to do it right.

Third, inaction carries a real and quantifiable cost. More on that next.

What Happens When Manufacturers Wait on MCP?

The survey data gives a specific answer: higher operating costs and competitive disadvantage, named by 56% of respondents as the direct consequence of delayed adoption.

But the longer-term picture is equally important. EVPs and VPs estimate that nearly one-third of their daily business processes will be automated or influenced by AI within three years. 

The manufacturers building that operating model now—on a foundation of connected, governed, structured product data—will automate from a position of strength. The ones who wait will spend their first year of implementation cleaning up data architecture and retrofitting governance, while competitors are already capturing the productivity gains.

How Does Propel MCP Address This?

Propel MCP connects Claude, ChatGPT, and other AI tools directly to live product records in Propel, through the same security and governance model your teams already rely on.

Every MCP interaction runs through Propel's enterprise trust layer, with zero data retention and full audit trails. Your IP is never stored or used to train external models. AI actions are governed by your existing user permissions, so agents only access what a given user is already authorized to see.

Critically, Propel MCP connects to actual live records. When an engineer asks about the status of an open change order, the answer reflects the current state of the record, not a version indexed last week. That distinction matters enormously in manufacturing, where decisions about supply chain, compliance, and product configuration are made on the assumption that the information is current.

And because Propel's unified product data model means product records, quality data, and compliance documentation are already structured and governed before any agent touches them, Propel MCP delivers accurate answers from day one. Learn more.

The Bottom Line

MCP has crossed from early adoption into strategic priority for manufacturing. The research is unambiguous: 91% of manufacturing professionals report increased organizational interest, 89% agree that connected product data is essential to AI's full potential, and 56% believe inaction will lead to higher costs or competitive disadvantage.

The manufacturers who will capture the full value of AI in the next three years are the ones building a connected, governed, structured product data foundation now. MCP is the access layer. PLM is the data source. And the time to connect them is before the pressure arrives.

See how Propel MCP connects your AI tools to live product data, with governance built in from day one. Request a demo today.

Frequently Asked Questions About MCP in Manufacturing

Q: What does MCP stand for?
A: MCP stands for Model Context Protocol. It is an open standard, created by Anthropic and soon adopted by other major AI tech leaders like OpenAI, Microsoft, and Google, that defines how AI agents connect to and act across enterprise systems through a single standardized interface.

Q: How is MCP different from a standard API?
A: APIs define how software communicates with a specific system. MCP defines a standard way for AI models to discover and use any MCP-compatible system consistently — without custom code for every new connection. The practical difference for manufacturers is significant: MCP turns a complex web of point-to-point integrations into a single governed protocol.

Q: Which enterprise systems can MCP connect to?
A: Any system with an MCP server can be connected. In manufacturing, the most common candidates are PLM, ERP, QMS, CRM, and procurement or supplier management platforms. Propel MCP supports inbound connections from AI clients like Claude and ChatGPT, and outbound connections from Propel’s agentic AI, Propel One, to systems like NetSuite, SAP, and SiliconExpert.

Q: Is PLM the most important system to connect through MCP for manufacturers?
A: According to Propel's survey of 400 manufacturing professionals, yes. PLM ranked first as the top business system for AI agents to access through MCP (49%) and first as the most valuable data source for AI actions (38%) — ahead of ERP, CRM, and QMS.

Q: What is the current state of MCP adoption in manufacturing?
A: MCP adoption is accelerating rapidly. In Propel's 2026 survey, 91% of manufacturing professionals report increased organizational interest, 96% of CxOs report increased interest, and over 90% have implemented or expect to implement MCP within 12 months.

Q: How does MCP relate to AI agents?
A: AI agents are software systems that can take autonomous actions in response to instructions. MCP is the protocol that gives those agents access to live enterprise data and the ability to act across systems. Without MCP, AI agents are limited to static knowledge or system-specific integrations. With MCP, agents can traverse your entire enterprise data landscape through a single governed interface.