Your ERP Knows What You Planned. Does It Know What Actually Happened?

Manufacturing technology is having an interesting moment.

Artificial intelligence is everywhere. Smart factories are moving from PowerPoint to production. Autonomous trucks are proving that automation’s biggest payoff often isn’t fewer people, it’s more hours out of the same asset. And proposed changes to North American trade rules could make the origin of individual components materially important to the economics of the finished product.

At the same time, Manufacturing Execution Systems are supposedly being disrupted, replaced, composed, moved to the cloud and reinvented.

Except there is a problem with the “MES is dead” narrative.

The global MES market was valued at roughly 17 to 18 billion dollars in 2025 and multiple industry analysts project it will more than double by the early 2030s, with cloud deployment and discrete manufacturing growing fastest of all.¹² Research from LNS Research points to the same story: sustained investment across the category, vendors pushing new cloud offerings, and large manufacturers increasingly looking for smaller, faster to deploy MES applications for what LNS calls the “rest of the fleet,” not just the flagship plants that already have full implementations.³

So MES isn’t disappearing.

The job it needs to  perform is changing.

And for many manufacturers, especially those already running a capable ERP such as Syspro, that raises a more useful question than: Do we need an MES?

The better question is:

Because that distance may determine how much value you get from everything you invest in next.

Welcome to the Last 50 Feet of Manufacturing

Your ERP is enormously important.

That is the last 50 feet.

And if what happens there is captured late, incorrectly or not at all, an uncomfortable thing happens.

Why Shop Floor Data Accuracy Is Becoming More Expensive

Historically, bad shop floor data created familiar problems.

• Missing inventory
• Stock adjustments
• Production delays
• Spreadsheet reconciliation
• Traceability headaches

A lot of walking around asking, has anyone seen pallet 2247?

Those problems haven’t disappeared. But three developments are making the consequences much larger.

1. Traceability Is Becoming an Economic Issue

Consider what is happening in the North American automotive market.

As part of the ongoing USMCA review, U.S. negotiators have proposed raising the required North American content in vehicles to as much as 82 percent, with half of that value produced specifically in the United States, up from the current 75 percent regional threshold.⁴ Detroit automakers have estimated the changes could add at least 2 billion dollars in annual costs per company, and General Motors alone has projected 2.5 to 3.5 billion dollars in gross tariff exposure this year.⁵

Forget the politics for a moment. Think about the operational implication.

A manufacturer may increasingly need to answer:

1. Which component did we use?
2. Who supplied it?
3. Where was it manufactured?
4. Which lot did it come from?
5. Which production order consumed it?
6. Which finished unit contains it?

That is no longer traceability for traceability’s sake.

It can affect cost. Margin. Compliance. Customer commitments. And potentially tariff treatment.

The chain of evidence eventually has to reach the factory floor. If the physical transaction and the digital transaction separate, the manufacturer may know what it intended to build without being able to prove exactly what it did build.

That is a very different business problem.

2. Automation Is Changing the Value of Operational Speed

There is another useful lesson coming from outside the factory.

Autonomous trucking is demonstrating that one of automation’s largest potential economic benefits is not simply replacing labor. It is increasing asset utilization. Federal hours of service rules cap a human driver at 11 hours behind the wheel in a 14 hour window. Executives at Volvo Autonomous Solutions have said an autonomous truck faces no such limit and can roughly double the productive hours generated from the same vehicle.⁶

The same economic idea applies inside a factory.

What happens when you can:

• Reduce waiting between production steps?
• Replenish material before the line stops?
• Find inventory without searching?
• Complete a changeover faster?
• Route a quality exception immediately?
• Remove manual re-entry between the floor and ERP?
• See work in progress as it actually moves?

The ROI isn’t always, how many people did we remove?

Sometimes the far more valuable question is:

That is asset velocity. And operational velocity becomes much harder when the system is constantly waiting for somebody to tell it what happened.

3. AI Needs Factory Truth

This may become the biggest issue of all.

Every manufacturer is going to hear increasingly impressive promises about industrial AI.

• Predictive scheduling
• Automated quality
• Intelligent replenishment
• Predictive maintenance
• Dynamic production planning
• AI agents coordinating manufacturing processes
• And some of those capabilities will be extraordinary

LNS Research has observed MES vendors pursuing two parallel paths: embedding AI directly inside manufacturing applications, and letting MES participate in broader enterprise AI workflows that span multiple systems, while cautioning that the technology is still early and evolving quickly.³

But AI has a fairly inconvenient dependency.

Imagine asking an AI system to optimize tomorrow’s production schedule.

• The ERP says you have 1,000 units of Component A. Reality says you have 847.
• The system says Lot 392 passed inspection. Reality says the paperwork is still sitting beside the workstation.
• The system says Work Order 1057 is 40 percent complete. Reality says the line stopped two hours ago.
• The system says the material is in Warehouse Location B17. Reality says someone moved it yesterday.
• The AI may produce a brilliant answer. To the wrong version of your factory.

That is why the foundational question for the AI powered factory isn’t:

Which AI should we buy?

It is:

Where MES Fits Alongside a Syspro ERP

The research from LNS Research provides an interesting view into this evolution.

The traditional MES model was largely about managing production inside the plant. That remains important. But LNS sees the category expanding in several directions: toward broader manufacturing operations management, industrial transformation, more composable applications, enterprise AI orchestration, and a renewed debate about where MES sits between enterprise IT and operational technology.³

There is another observation worth paying attention to.

Large manufacturers don’t necessarily want to deploy enormous, complex MES implementations at every facility. LNS describes growing interest in serving the rest of the fleet with smaller, easier to implement applications rather than treating every plant like an aircraft carrier, and notes that vendors are increasingly breaking MES into more digestible, composable applications manufacturers can adopt incrementally.³

That makes sense.

Most manufacturers don’t wake up thinking ‘we need a giant new software category’.

They wake up thinking:

• Our inventory is wrong
• Our traceability is painful
• Receiving takes too long
• We can’t see work in progress
• Quality information arrives too late
• Operators hate the system
• We’re entering the same data twice

Solve that problem. Then solve the next one.

The Human Side of Shop Floor Software

This may be the most overlooked part of manufacturing transformation.

LNS Research argues that traditional MES vendors often underestimate the role of the connected frontline worker.

An operator doesn’t spend the day thinking about MES architecture. They are trying to win the shift. They may be running equipment, moving materials, changing jobs, solving problems, completing inspections and responding to interruptions.

Software that requires six screens, unnecessary data entry and weeks of training can actually slow the very operation it was supposed to improve.

The technology has to meet the operator where the work occurs. That means the transaction should be simple. Immediate. Relevant. And ideally difficult to get wrong.

That is a design philosophy we believe strongly at TransLution.

Where TransLution Fits: Syspro MES Integration Built for the Floor

For manufacturers using Syspro, we don’t believe the answer is replacing a powerful ERP.

Quite the opposite. We want to make the PRO better.

Syspro understands the business. TransLution extends that intelligence to the physical point of execution.

TransLution integrates with Syspro while providing manufacturing, warehouse and quality workflows designed around what operators actually do: receiving, put away, transfers, production, material consumption, quality capture, stock movements, lot and serial tracking and other floor level transactions. Our own customer implementations include real time Syspro updates, warehouse control, production visibility and RFID driven work in progress tracking. See how manufacturers are putting this to work in our case studies.

The objective is simple.

Because when they agree, something much bigger happens.

• Your inventory becomes more trustworthy
• Your planning becomes more useful
• Your traceability becomes more defensible
• Your operators spend less time fighting systems
• Your equipment spends more time producing

And the AI systems you deploy tomorrow finally have a reliable representation of the factory they are supposed to optimize.

The New Manufacturing Technology Stack Starts With Truth

There will be plenty of debate over what we call the next generation of manufacturing systems.

• MES.
• MOM.
• Connected worker.
• Composable manufacturing applications.
• Industrial AI.
• Smart factory.

The terminology will keep changing. The underlying requirement won’t.

Every sophisticated system above the factory floor eventually depends on knowing:

1. What arrived?
2. Where is it?
3. Who owns it?
4. Has it passed inspection?
5. What was consumed?
6. What was produced?
7. Where did it go?
8. Can we prove it?

Call that MES. Call it manufacturing execution. Call it the connected frontline. We prefer something simpler.

And before manufacturers spend millions adding more intelligence above the factory, it may be worth making sure the systems they already have know what is actually happening below them.

Five Questions to Ask Before Your Next Manufacturing Technology Investment

Before adding another layer of software, automation or AI, ask:

1. Does our ERP know what is physically happening on the floor right now, or what someone told it happened later?
2. Can we trace a finished product back through the exact materials, lots, inspections and production transactions that created it?
3. How much operator time is spent entering, finding, correcting or reconciling information rather than producing value?
4. Could we increase output from our existing people, equipment and inventory simply by removing waiting, searching and transaction friction?
5. If we gave an AI system our operational data tomorrow, would we trust it enough to let that AI optimize production?

If any of those questions make you uncomfortable, the next step may not be replacing your ERP.

It may be closing the last 50 feet between Syspro and the work.

And when the two agree, everything you build on top gets smarter.

Frequently Asked Questions

Sources

1. Manufacturing Execution System (MES) Market, Mordor Intelligence, 2026. https://www.mordorintelligence.com/industry-reports/manufacturing-execution-systems-market ↩
2. Manufacturing Execution System (MES) Market Expected to Reach US$44.67 Billion at 13.3% CAGR by 2034, GlobeNewswire, June 2026. https://www.globenewswire.com/news-release/2026/06/29/3318698/0/en/Manufacturing-Execution-System-MES-Market-Expected-to-Reach-US-44-67-Billion-at-13-3-CAGR-by-2034.html ↩
3. Early Observations on MES: The King is Dead, Long Live the King, LNS Research. https://blog.lnsresearch.com/early-observations-on-mes-the-king-is-dead-long-live-the-king ↩ ↩2 ↩3 ↩4
4. U.S. Pushes Tougher American Content Rules for Cars in USMCA Review, The Detroit News, May 2026. https://www.detroitnews.com/story/business/autos/2026/05/29/u-s-wants-much-more-american-content-in-cars-as-usmca-talks-begin/90313608007/ ↩
5. USMCA Revisions Could Add At Least $2 Billion in Annual Costs, CBT News, citing Reuters, August 2026. https://www.cbtnews.com/revised-usmca-proposals-could-cost-billions ↩
6. Volvo Autonomous Solutions Lays Out Driverless Trucking Roadmap, ACT News, June 2026. https://www.act-news.com/news/volvo-autonomous-solutions-lays-out-driverless-trucking-roadmap/ ↩