It’s one of the most common refrains in manufacturing: “The ERP doesn’t work.” Teams go live, expectations are high, and then reality sets in. Reports are wrong. Inventory numbers don’t add up. The operations team is working around the system instead of in it. And somewhere along the way, someone decides the software is the problem.
Usually, it isn’t.
The Real Culprit: Process
Business Central, SAP, Oracle, NAV: these are mature, globally deployed platforms used by tens of thousands of manufacturers every day. When implementations go sideways, the root cause is almost never a software defect. It’s a process problem that the software faithfully and brutally exposes.
Manufacturing companies have often spent years developing workflows built around paper, spreadsheets, and institutional knowledge. Those workflows have workarounds baked in, informal steps which aren’t documented, and inefficiencies that everyone has simply accepted. When an ERP implementation begins, the pressure is on to go live, and the path of least resistance is to replicate the old process inside the new system.
That’s the lift-and-shift trap, and it’s where a lot of implementations go wrong.
Moving a broken or inefficient process into a new ERP doesn’t fix the process. It just gives you a more expensive version of what you already had and often a slower, more confusing one, because now your team has to learn a new interface on top of navigating the same underlying problems.
The right sequence is process first, technology second, automation third. Get the business process right. Then configure the software to support that process. Then build automation on top of clean, reliable data. Skipping or shortchanging the first step undermines everything that follows.
What “Second Opinion” Engagements Reveal
Manufacturers who reach out for help after a troubled go-live often can’t pinpoint exactly what went wrong. Things just feel off. Numbers don’t reconcile. Teams have lost confidence in the data.
When you dig in, the patterns are consistent. Inventory costing is frequently a trouble spot because it’s technically complex and easy to misconfigure in ways that compound over time. Data accuracy problems show up across the board. And almost always, somewhere in the operation, people are working outside the system: a spreadsheet for BOMs, a whiteboard for scheduling, a manual workaround for something the team decided was “too hard” to do in the ERP.
Each of these workarounds is a symptom. The diagnosis is usually that the process wasn’t mapped carefully enough before the implementation began, or that user adoption was treated as a training event rather than an ongoing effort.
The Data Foundation Problem
One of the most overlooked prerequisites for a successful ERP go-live is clean data and one of the most common mistakes is underestimating how much work that requires.
For manufacturers moving into structured production management for the first time, or migrating from a legacy system where certain modules were never used, the data preparation work can feel overwhelming. Bill of materials that lived in spreadsheets need to be validated and structured. Item records need to be accurate and complete. Costing methods need to be decided before data is loaded, not after.
A practical approach is to start narrow. The 80/20 rule applies here: identify the items and assemblies that account for the majority of your production volume and get those right first. A clean, accurate foundation for your top-volume products will deliver more immediate value than a rushed, incomplete migration of everything at once.
Without good data, everything downstream is compromised. Production orders, planning, costing and purchasing all run on the integrity of the underlying item and BOM data. There’s no shortcut around this step.
User Adoption Is Part of the Implementation
There’s a growing trend toward faster, lower-cost ERP implementations — packaged approaches that compress the timeline and reduce consulting hours. These can work well for companies with straightforward operations and teams who are comfortable learning in a self-directed way.
But speed has tradeoffs. The piece which often gets compressed in a fast implementation is the hand-holding that drives genuine user adoption. ERP screens are dense. Most fields on any given screen won’t apply to a particular user’s role. Without intentional guidance on what matters and why, users get overwhelmed, start working around the system, and data quality degrades from day one.
Implementation success isn’t just about going live on time. It’s about whether the team is actually using the system correctly six months later. That requires setting realistic expectations. Implementations are hard, and pretending otherwise doesn’t help anyone while also investing in the user experience enough to make adoption stick.
Change is uncomfortable. People default to familiar habits. Acknowledging that reality and building a plan to work through it is part of a responsible implementation, not a nice-to-have.
The Bottom Line
ERP implementations don’t have to be the organizational nightmares they’re often made out to be. But they do require honest assessment of your current processes before the software ever enters the picture, a commitment to data quality that most teams underestimate, and a genuine investment in making sure the people using the system are set up to succeed.
The software works. The question is whether your organization is ready to work with it.
CADTALK helps manufacturing companies streamline the data handoff between engineering and ERP, reducing the manual work that undermines data quality and slows production. Schedule Your Personalized CADTALK Demo Today | CADTALK
