And Where Each One Stops.

Most manufacturers have already tried to fix the shop floor. They didn't sit around waiting for someone to hand them a solution. They ran projects. They spent budget. They brought in consultants. And in a lot of cases, those projects are still technically running, just not producing the results anyone expected.

We talk to manufacturers across aerospace, defense, medical devices, electronics, and industrial production, and the same three approaches keep coming up. Extend the ERP. Go paperless. Build something internally. We've written about what's actually broken on the shop floor. This post looks at what happens when manufacturers try to address it, and why so many of those attempts fall short of what a manufacturing execution system (MES) provides.

Attempt 1: Extending the ERP

Why ERP manufacturing modules fall short on the shop floor

On paper, it makes sense. Your ERP has the work orders, the BOM, the purchase history. It knows what's planned. The vendor has a manufacturing module. Why not use it?

We've seen this play out enough times to recognize the pattern. The demo is clean, the module works exactly as advertised, and then it meets an actual production floor.

When the demo doesn't survive contact with reality

One manufacturer walked us through what happened after their rollout. The module's assumptions about how production moved didn't match how work actually moved through their facility. Time entries couldn't be corrected after the fact. Material consumption posted to the wrong accounts. Every fix they applied created a new workaround, and the workarounds added more overhead than the module was eliminating. After several months, they went back to paper travelers. The project left behind a system nobody used and a paper process that had never actually gone away.

A second manufacturer got further. Their module could open and close work orders. What it couldn't do was deliver work instructions at the operator level, enforce process sequence, flag a certification problem before a build started, or track quality data against the specific unit being built. Paper forms kept running in parallel. The module added process without taking any away.

ERP does what it was designed to do. It runs the business side of production: orders, material requirements, purchasing, and cost. Production execution is a different job. Knowing that a work order opened and closed is a different problem than knowing who built what, in what sequence, against which revision, with which lot of material, and whether anything deviated from the plan in the process. That second set of problems is what an MES is designed to handle. ERP planning also stops short of the floor. It sets what to order and when it's due, while deciding which job runs on which line, in what order, around which constraints, takes production planning and scheduling built for that level of detail.

ERP and MES work best side by side, each handling the job it was built for. For more on how ERP and MES are actually designed to work together, see the benefits of connecting ERP and MES.

Attempt 2: Digitizing the Records Without Changing the Process

When moving to tablets doesn't move the needle

Go paperless. It sounds obvious once someone says it out loud. Paper is slow, it gets lost, it can't be searched. Put screens and tablets at every station and it feels like you suddenly have data. The records are cleaner. Audit prep should be easier. Quality problems should be easier to track.

Cleaner records help, though on their own they rarely change quality outcomes.

Where paperless tools stop

One manufacturer had rolled out digital inspection forms across their production line. Operators were filling them out on tablets, data was being captured, the paper was gone. But audit prep was taking just as long as it always had, because the system couldn't automatically generate the compliance reports their customers required. Someone was still manually pulling data and assembling reports by hand. The records were digital, and the reporting work was exactly the same as before.

Another manufacturer moved work instructions from printed binders to PDFs on screens at each station. Always the current revision, always visible. That was a real step forward, and for a lot of manufacturers it's the right place to start. What the PDFs couldn't do was confirm the operator was certified for the task, verify that the right materials were staged, or stop someone from moving to the next step before the previous one was signed off. Getting those controls doesn't mean replacing everything at once. Manufacturers who go paperless can add tracking and data collection to the same work instructions, then layer in certification checks and step enforcement as their needs grow. That depends on the paperless system they pick. Some document viewers have no way to add tracking or enforcement later, so it's worth asking about the path forward before choosing one.

There's a difference between recording work and governing it

The distinction that tends to get missed here is between documenting execution and actually controlling it. When a defect gets caught three operations downstream, the root cause is usually timing. Inspection was chasing production rather than running alongside it. Quality management built into the execution flow catches problems before they move. Quality inspection bolted on after the fact catches them later, in a cleaner database, but still after.

In practice, digitizing the records doesn't change a few things that matter quite a bit:

• An operator can skip a step or complete steps out of sequence, and nothing flags it
• The work instruction on the screen isn't tied to the specific unit being built, it's the general version
• Certification gets logged after the work, not confirmed before it starts
• Compliance documentation still has to be assembled manually, it just comes from a digital source now

Paperless tools can show operators the work and record it afterward. Changing the conditions under which the work happens takes execution controls.

Attempt 3: Building Something In-House

Why homegrown manufacturing tools outlive their usefulness

Sometimes the most honest thing we hear in a discovery conversation is a manufacturing engineer saying, with real pride, "we built our own system." And they're usually right to be proud of it. These tools get built because someone understood the problem better than any vendor did, and they solved it. They just solved the version of the problem that existed when they built it.

The knowledge problem

One manufacturer described their internal tracking system and how well it worked, right up until something changed. A new product line. A customer requirement the system's fields couldn't accommodate. A failure mode someone needed to track differently. Each change meant going back into the code, being careful not to break something else, relying on the person who originally wrote it to remember what they were thinking. That person was still there. Nobody had a real plan for what happened when they weren't.

The growth problem

A second manufacturer had something more sophisticated. Work-in-process tracking that was genuinely reliable. But it couldn't enforce process sequence, manage operator certifications, connect to the BOM in real time, or produce traceability records that customers were starting to require as standard. It had been built for a company that no longer existed, and the company that existed now had outgrown it.

Nobody designs a homegrown tool for a company that will be three times its current size, operating at a facility the original developer has never seen, under a compliance regime that didn't exist when the tool was built. The reckoning usually comes from outside the team that built it. Sometimes it's a customer audit. Sometimes it's a regulation with a hard deadline. A lot of the time it's someone leaving, and the next person inheriting a system they can barely maintain and can't modify without breaking something else.

The tool gets defended long past its useful life because the credibility of the person who built it is attached to it. That's a dynamic that's hard to change from inside.

Deloitte cites research showing workers lose the equivalent of 32 days a year switching between applications just to find the information they need. Homegrown tools cut that overhead in the workflows they were built for. Everything else stays as it was.

What All Three Have in Common: No MES

The missing piece is an execution layer between the plan and the floor

Each of these approaches improved something. The ERP module gave the front office better visibility into orders. Digital records made the paperwork searchable. Homegrown tools eliminated manual steps in specific workflows. None of that is nothing. But the relationship between the floor and the system didn't change in any of them.

Without a purpose-built execution layer, the operator is still the one carrying the information. They get an instruction, do the work, report back. The system doesn't show up until after the fact, if it shows up at all. It has no way to influence what's actually happening during the work. Whether it captures what happened depends entirely on whether the operator filled something in.

What a manufacturing execution system does is move the system into the work itself rather than after it. When an operator scans into a station, the system checks whether they're certified before anything starts. It surfaces the work instruction for this specific unit and revision. If a prior step hasn't been signed off, the work doesn't start. When it's done, the record updates. The operator didn't do any of that manually. It just happened as part of doing the job.

Compliance records, traceability data, and quality documentation accumulate as a byproduct of the work. Operators don't have to complete them as a separate task. That is the core difference an MES makes, and a better form or spreadsheet can't produce it.

That execution layer doesn't have to arrive all at once. Many manufacturers start with digital work instructions, add tracking and data collection at the stations that need it, and bring in certification checks, material verification, and step enforcement as production demands them. Each piece builds on the same work instructions and the same records. This only works when the first piece was built to take the next one. Manufacturers who outgrow a standalone viewer can end up re-creating their instructions in a new system, which is the kind of restart a staged approach is supposed to avoid.

Where This Leaves Most Manufacturers

The operations we see getting this right tend to have one thing in common: they got specific about what the problem actually was before they went looking for a solution. They moved past broad statements like "our floor visibility is poor" or "we need to go digital" to something more precise, like: the system has no presence when the work is happening, so we have no ability to enforce anything or catch problems before they move downstream. That is the problem MES software is built to solve.

The manufacturers who've been through one of the three approaches above often get there faster. They know the ERP module wasn't the answer. They know cleaner records didn't change the outcomes. They know the internal tool hit a wall. That experience is worth something. It's usually what produces a clear enough picture of the actual problem to do something about it.

FactoryLogix is built to be the MES execution layer that ERP modules, paperless tools, and homegrown tools can't provide.

See how it works in production environments like yours: Explore FactoryLogix or request a demo.