Why most manufacturing system projects stall

Choosing the system is the easy half. Here is why the project so often fails to land, and what changes that.

H² Consulting · 27 October 2026 · 4 min read

You have seen it happen, or heard about it from someone who has. A manufacturer picks a shiny new ERP, spends a serious sum, and two years later half the team is back on spreadsheets working around it. The money is gone and the problems are still there. This is the most common way system projects fail, and it almost never fails for the reason people think.

The decision is not the finish line

Choosing a system feels like the hard part, so firms pour their energy into selection and treat go-live as the end. It is the beginning. The real work is getting your actual processes and data into the thing, and getting your people to use it.

Why good projects stall

Three things sink most of them. The requirements were thin, so the system never fit how you really make things. The data was migrated dirty, so people stopped trusting it. And nobody stayed to embed the change, so the team quietly drifted back to the old way the moment attention moved on.

• Thin requirements. The system was chosen on a demo, not on how your shop floor actually runs.
• Dirty data. Garbage in, garbage out. One bad stock figure and people stop believing the screen.
• No one owns the landing. Training was a half-day, and the moment the project team left, habits snapped back.

What makes it stick

This is exactly why we do not hand over a plan and leave. We define the requirements properly, clean the data before it moves, train people on how their own work changes, and stay through go-live until the new way of working is simply the way things are done. A system that gets used is worth more than ten that get bought.