Five mistakes companies make when implementing ERP

1. Digitising the existing process as-is

This is the most common mistake. The company wants its paper or spreadsheet process translated one-to-one into software. The result: the same clumsiness, now on a screen. The real value of ERP is the chance to review the process. If nobody can explain why a step exists, that step is probably the residue of a problem solved years ago.

The right approach is to question every step: what would happen if this step didn't exist? If the answer is "nothing", it goes. If the answer is "we'd lose that control", it stays, but perhaps automated.

2. Trying to please everyone

In scoping meetings every department states its wish and the list swells. Six months later you have a system nobody quite likes and everyone uses a fragment of. This is the most expensive ERP mistake because it is hard to undo.

The cure is ruthless prioritisation: the first release solves only the points where the operation stalls. Everything else goes into a second phase, and that phase genuinely arrives. Once people start using the system the wish list changes on its own, because they finally know what they need.

3. Leaving data migration until the end

The software is finished, it is time to move the data, and the project stops there. Half the account records in the old system are incomplete, stock codes are inconsistent, the same customer is registered under three different names. That clean-up takes weeks, and usually nobody budgeted time for it.

Data clean-up must run in parallel with development. Start it in the first week of the project, not the last.

4. Treating training as a one-day event

After a single group training session everyone is assumed to be able to use the system. In reality people learn while doing their own work. Having someone continuously reachable for the first two weeks is far more effective than one full day of training.

Naming a "super user" per department also helps. That person gets deeper training and becomes the first point of contact for colleagues, so not every question reaches the development team.

5. Forgetting to measure

ERP is live, everyone uses it, the project is done. But what changed? Did the time to process an order drop? Did stock count discrepancies fall? Did the share of overdue receivables decline?

Before starting, pick three to five numbers and write down their current values. Look at the same numbers six months later. This shows both the return on the investment and where the next phase should go.

• Average time from order entry to dispatch
• Discrepancy rate in stock counts
• Overdue receivables as a share of total receivables
• Time to close the month
• Weekly hours spent on manual data entry