Do you hate your ERP?
It’s not you. It’s the model layer.

If the ERP feels like the enemy, you are in a large and reasonable majority. The useful question is not whether the frustration is justified — it is — but where it actually comes from. Almost always it is aimed at the wrong place: at the people who use the system, or at the software itself. The real culprit is the layer between them.

You are not imagining it

The dissatisfaction is measured, not anecdotal. Roughly two in three mid-market manufacturers report being unhappy with their ERP. Between 55% and 75% of ERP projects miss their stated objectives. About half of the people living inside a rollout will call it a failure. When a number is that consistent across that many companies, it stops being a story about any one team’s competence.

So set the blame down. It is not the controller who built a workaround in Excel. It is not the scheduler who stopped trusting the on-time report. It is not even, usually, the ERP as a piece of software — these systems run enormous, complicated manufacturers every day. The malfunction is somewhere more specific.

Separate the people from the malfunction

Most ERP hatred is really frustration with the model layer — the definitions, the costing rules, the date logic, the extracts and refresh paths that sit between the raw transactions and the report someone quotes in a meeting. When margin comes out three different ways, when the delivery number can’t be trusted, when a refresh breaks and no one can fix it, the ERP gets the blame. But those are model-layer failures, and they are fixable without ripping out the system that runs the plant. We take apart the anatomy of them in Why your reports disagree.

This distinction matters because it changes what you do next. If the software is the enemy, the only move is a wrenching, expensive replacement. If the model layer is the problem, the move is smaller, cheaper, and far more certain: reach the data, define the numbers once, and build the reporting on definitions you own. Same system, trustworthy numbers.

Sometimes the ERP really is the problem

Not always, though. Sometimes the system genuinely no longer fits — a business has outgrown QuickBooks and a wall of spreadsheets, or it is limping along on something homegrown, unsupported, or acquired in a deal and never integrated. When that is the honest diagnosis, we convert manufacturers onto the two ERP stacks we know deepest, Infor CSI/Syteline and Global Shop Solutions:

One line worth stating plainly, because it is our discipline and not a limitation we hide: we do transformation and analytics on any ERP, but we take on ERP implementation work only on those two stacks. We would rather do a conversion we can stand behind than sell one we can’t.

Which problem you have is the audit’s job to say

Model layer or the system itself — you should not have to guess, and you should not take the word of anyone with something to sell on the other side of the answer. That is what the audit settles. It grades each ERP module — healthy, misconfigured, or abandoned — and tells you honestly whether the fix is a model built on the system you already run, or a conversion to one that fits. Both sides learn the same thing at the same time, which is the only fair way to start.

If you have been quietly furious at your ERP for years, the most useful first step is not a new module or a new hire. It is two weeks and a graded report that finally names the real problem — and, more often than the frustration would predict, absolves the software.