What is an MES, and does a small shop need one?

Somebody at a trade show or a software demo told you that you need an MES. You nodded. Then you got back to the shop and realized you could not explain what one is, what it would replace, or why your ERP does not already do it. This post covers what a manufacturing execution system does, where the standard definitions come from, how it differs from an ERP, and how to tell whether a shop your size needs one. It ends with a checklist you can run on Monday.

What an MES is, in plain words

A manufacturing execution system is the software that runs the shop floor between the moment a job is released and the moment it ships. It tells people what to run next, records what actually happened, and shows everyone the same status while the work is still in progress.

The standard home for that definition is ISA-95, the enterprise-control integration standard from the International Society of Automation. ISA-95 sorts a manufacturing business into five levels, from the physical process up to business planning. Level 3 is "manufacturing operations management," and the standard names manufacturing execution systems as the systems that live there. Level 4 is business planning and logistics, which is where ERP lives.^[1]

Levels in the table are from ISA's own summary of the model.^[1] The shop-floor examples are mine.

A peer-reviewed description of MES puts it this way: the MES binds the upper level (ERP) with the lower level (the factory floor), and in many companies that bridge is delegated to a human activity.^[2] If someone in your shop spends the morning walking the floor to find out where jobs are and then typing the answers into a spreadsheet, that person is your MES.

The standard list of MES functions

MESA International, the trade association that named the category, published a list of eleven MES functions in the late 1990s and has updated it since. The list is a union of everything an MES may do. A plant does not have to implement all eleven to get value, and which ones matter depends on the industry and the level of automation.^[2]

Here is the list, with my translation for a job shop.

The function names and short definitions come from MESA's list as reproduced in the peer-reviewed literature.^[2]

Notice what is on that list and what is not. There is no quoting, no accounts receivable, no general ledger. Those are ERP functions. The MES list is about executing work that has already been sold.

How an MES differs from an ERP

The cleanest way to separate the two is time. The OPC Foundation's summary of ISA-95 says Level 3 systems typically operate on time frames of days, shifts, hours, minutes, and seconds, while Level 4 systems typically operate on time frames of months, weeks, and days. The same summary notes that each level has its own definition of "real time." For Level 3 it means information available a few seconds after a shop floor event. For Level 4 it means logistics and material information available daily or within a few hours after the end of a shift.^[3]

That difference explains a lot of the frustration owners feel after buying an ERP and expecting to see their floor. The ERP was built to answer "what did we ship this month and what did it cost." It was not built to answer "which job at the press brake is going to make the Friday truck." They are different systems because they run on different clocks.

The two systems are meant to talk to each other. ISA-95 exists mostly to define the interface between Level 3 and Level 4.^[1] NIST's standards review calls ISA 95 a commonly used reference model for automated interfaces between enterprise and control systems, and notes that the B2MML format published by MESA is a widely adopted implementation of its data models for connecting ERP with MES.^[4] An ISA committee leader who helped write the standard describes the levels as logical boundaries that let a practitioner define where one system stops and another starts.^[5]

The longer version of that argument is in You don't need a new ERP. You need to see your floor.

Does a small shop need one?

Not every shop does, and nobody can tell you from the outside.

The case for MES in small shops has been made by people with no software to sell. A 2021 post on the NIST Manufacturing Extension Partnership blog, written by an MEP center operational excellence manager, argues that small and medium manufacturers are often resource- and talent-challenged, and that this is exactly why an MES is attractive to them. He describes an MES as manufacturing operations management software that accounts for inputs and outputs, says it works like a real-time audit of operations, and notes it can stand alone as a scheduling or quality system and feed production data into the ERP.^[6] On cost, his estimate is that cloud-based options start at roughly the cost of one full-time equivalent.^[6] Treat that as one practitioner's number.

A Georgia MEP article from 2024 adds a method. It says many small and mid-size manufacturers are not familiar with MES, and that they hold back because they do not know the right solution, worry about cost and return, and worry about support after go-live. Its advice: first identify where you collect data today (whiteboards, spreadsheets, who collects it, what happens to it), then start small with the one issue that has the most impact. Many manufacturers start with a module that captures downtime.^[7]

The research points the same direction, with a caution. A University of California, Irvine survey of 54 owners and employees of US small and medium manufacturers, recruited through the MEP network, found the top barriers to adopting new production technology were not having the time (26%) or the money (22%). The same survey found that "having to do tasks manually" and "having to wait for information" were the problems most often reported as hurting productivity.^[8] The authors conclude that small manufacturers do better targeting limited resources at specific technologies for specific needs rather than buying the whole smart-factory menu.^[8]

A 2024 study of MES adoption in small and medium manufacturers notes that MES has existed since the 1990s, that the concept was already unclear to many manufacturers before Industry 4.0 made it more so, and that smaller companies know they need to move but struggle with when and where to start.^[9] If that describes you, you are not behind. You are normal.

What I have seen

In my shops, the question was never "do we need an MES." The question was "why is this order late and why did nobody know until the customer called." In 13 years of fixing late orders by walking the floor and asking why, the pattern was the same every time: good people, scattered data, and nobody looking at the same numbers. One steel products maker I worked with had over $700,000 tied up in late orders. Six months later it was under $50,000. No new system. Everyone just looked at the same numbers every day.

That experience is why I am careful about this topic. A shared number on a whiteboard can beat a lot of software. The Lean Enterprise Institute defines visual management as placing tools, parts, production activities, and performance indicators in plain view so the status of the system can be understood at a glance by everyone involved.^[10] That is the job. An MES does that job at a scale and speed a whiteboard cannot keep up with. It is not the only way, and it is not the first way.

Off-the-shelf or built to fit

Once a shop decides it needs one, the next fork is whether to buy a packaged MES or have one built around the shop. Both are legitimate.

The case for off-the-shelf

A packaged MES exists today, has been debugged by other customers, and comes with a vendor who has implemented it before. For a shop whose routings look like everyone else's, that is a real advantage. The trade-offs are the ones a Georgia Tech research team described when they set out to build a low-cost alternative: a commercial MES typically arrives alongside an ERP, the pair can be a large expense that small manufacturers cannot always afford, deployment can require dedicated hardware, and some systems do not address what individual employees need to know.^[11] Per-user licensing is the other quiet cost: it punishes you for putting a screen at every work center, which is the entire point.

The case for built to fit

The same team showed that a working MES can be built from free and open-source tools, run on ordinary Android devices, store data in the cloud, and pull machine data from CNC controls through an open standard. They judged the approach particularly suitable for small manufacturing enterprises because it is low-cost and easy to implement.^[11] That was a research project, not a product, but it makes the point: the hard part of an MES is not the technology. The hard part is matching it to how your shop actually routes work.

That is the approach HixonOps takes. We build the MES around the shop's own routings and work centers, connect it to the ERP and CRM already in place rather than replacing them, deploy it in the customer's own cloud account, and charge a flat license, never per user. You can read how that works at /#mes.

How to choose fairly

Ask the same questions of both options:

• Does it dispatch from my routings as they exist, or do I have to change how I route work to fit the software?
• Where does the data live, and can I get it out without the vendor?
• What does it cost when I put a tablet at every work center?
• Can it start with one function (job status, downtime, or scrap at the operation) and add the rest later?
• Who maintains it in year three?

A packaged system that answers those well is a good buy. A custom build that answers them badly is a bad build. The label does not decide it.

Start smaller than you think

Whatever you pick, the sequence matters more than the product. The MEP guidance is consistent: understand how you collect data now, pick the one problem with the most impact, and start there.^[7] The UCI survey adds a second reason to go small: the top barriers for shops like yours are time and money, and a narrow first step spends less of both.^[8]

A first step that costs nothing: for two weeks, have every operation record start time, stop time, good count, and scrap count on the traveler. Then add up, by work center, how much time jobs sat waiting versus running. That number tells you whether you have a visibility problem or a capacity problem, and which MES function you need first. In my experience it is usually visibility. The capacity was there. Nobody could see it.

The bottom line

An MES is the Level 3 system: it runs the floor between release and ship, on a clock measured in minutes, and it is defined by standards bodies with no software to sell you.^[1][3] An ERP runs the business on a clock measured in weeks.^[3] Most small shops already have an MES. It is a person with a clipboard, and the shop depends on that person more than anyone admits.

Whether you need software for that job depends on how many of the ten signs above are true in your building. If it is most of them, the answer is yes. Then the only remaining question is whether the system should bend to your shop or your shop should bend to the system. Ask it out loud before you sign anything.