Ten ERP mistakes small manufacturers keep making

I have spent a good part of my working life helping small manufacturers choose and implement ERP and MRP systems, first as a software consultant and now through Smart Manufacture. The businesses change. The mistakes do not. The same ten come up again and again, and every one of them is avoidable.

A note on where we stand before we start. We are a UK implementation partner for MRPeasy and we may earn a referral fee if you buy it through us. None of what follows depends on which system you pick. Get these ten things wrong and the best software in the world will not save you.

1. Writing the requirements for one department

It sounds too obvious to mention, and yet it is the most common mistake of all. A manufacturer picks a system to fix its most painful problem, or lets one department choose it, and ends up with software that suits the shop floor and ignores purchasing, stock and sales. Sometimes a new manager simply brings in whatever they used at their last company without checking it fits this one.

Get people from every part of the business round a table and write down what each of them needs before you look at a single demo. Oceanic Saunas in Wolverhampton spent the best part of a year working out their requirements before choosing, and Director Sarah Jones says that homework is what gave her the confidence to recognise good software when she saw it.

2. Buying for this year, not the next ten

Your current requirements are hard enough to pin down. Your future ones matter just as much. A manufacturing system should last you 10 to 15 years, and a basic package that fixes today’s problem can run out of road remarkably quickly.

I have seen companies outgrow an ERP in under two years and have to start the whole implementation again. That is money down the drain. Look for a system that can scale with you, that integrates with third party apps for the functions it does not cover, and whose developers actually listen to customer suggestions.

3. Having something built from scratch

Most small manufacturers believe their requirements are unique. Having worked with a great many of them, I can tell you the processes and the problems are remarkably similar from one to the next. Despite that, plenty go down the custom software route, and it costs tens or hundreds of thousands of pounds more than it needs to.

Two further problems. Software developers rarely understand manufacturing, so the result tends to miss the detail that makes a production system work. And every bug fix or new feature depends on the developer being available and willing. It is not unusual for a bespoke system to lose its support entirely within a few years.

4. Expecting a quick fix

Implementing a system that touches every part of the business is not supposed to be easy. Even the most user friendly systems, and even with a consultant doing most of the work, take real effort to set up properly.

Manufacturers with no previous ERP experience often expect a magic wand. They do not free up the people or the time, they rush the set up without testing or validating data, and then the “real work” takes over and the project loses momentum. The result is a system nobody uses, a quiet return to the old spreadsheets, and a wasted budget.

Treat the software as a structure that lets the business fix its own problems, not a cure. Go in stages, usually starting with inventory, and move on when each area is working. Set a few realistic short term measures so you can see the return as you go.

5. Migrating bad data

Rubbish in, rubbish out. Poor data guarantees poor results, and clean data is the foundation of everything the system will do for you. The trap is assuming that copying everything across from the old system is good enough. Often that legacy data is out of date, inconsistent, in the wrong format, or built around the old way of working that you are trying to leave behind.

Review the quality and relevance of your data before it goes anywhere near the new system. Starting fresh is always better than building on a bad foundation. Oceanic Saunas found their SKU list had been created by too many people over too many years and cleaned it up completely before go live. It was the right call.

6. Nobody in charge of the project

Every implementation needs a sponsor who can drive it forward and clear obstacles out of the way. That person needs credibility across the business, real authority, and a deep understanding of how the operation works. Give the job to someone without those things and the outcome suffers.

I have seen a health and safety manager with little knowledge of production put in charge of an ERP project. The team did not respect the choice, the workforce did not buy in, and resistance set in from day one. Choose someone who will use the system daily and who can hold people accountable.

7. Managers who were never asked

The heads of each function need to be behind the project, and that is usually treated as an afterthought. Without them you get resistance to change and a system that is only half used.

Identify your key stakeholders at the start, involve them in the decisions that affect their area, and keep them updated on progress. A disengaged manager can derail the whole thing, so their support has to be there from beginning to end.

8. A workforce that was told, not involved

New technology changes how people do their jobs, and that is unsettling. Handle it badly and the natural resistance turns into real pushback.

Clear, consistent, two way communication is the answer. Deal with the fears directly, whether that is worry about being replaced or simply about how the day will change. Explain what the system will make easier. Involve people in the set up and support them through it, and the sceptics often become your strongest advocates. Sarah Jones at Oceanic told us her staff now “almost compete to solve issues” in the system. They did not start out that way.

9. Skimping on training

The best system in the world fails if people cannot use it. Small manufacturers routinely underestimate how much training is needed, and how much of it needs to be ongoing rather than a single session at go live.

Training should cover not just which buttons to press but how each team’s day to day process changes. Explain the why, give people hands on practice, and they will use the system with confidence. Hovat, a packaging manufacturer we worked with, brought staff into the test system gradually before the switch, and one of their production team is now the go to person whenever anyone needs help.

10. Leaving the accounts package on its own

An ERP is meant to be the hub of the business, but it does not work in isolation. Leave it disconnected from your accounting software and you get duplicate entry, errors, and a lot of missed automation.

Before you implement, list every system the ERP needs to talk to and plan those connections in. Vanquish Hardware Protection moved from Sage to Xero at the same time as going live on MRPeasy, precisely because the two integrate natively, and their order processing and cost tracking now flow through without rekeying.

Avoiding all ten

None of this is complicated. It is a matter of doing the boring parts properly: the requirements, the data, the people, the training. If you would like a second opinion before you commit to a system, or help getting one live, get in touch. You can also read how other UK manufacturers have handled the move.