BlogChoosing an ERP

Why ERP projects fail, and how to avoid it.

ERP projects rarely fail because of the software itself. They fail because scope, data and people are left until the end.

1. Everything goes live at once

The most common mistake is a big-bang start: sales, warehouse, production and invoicing on the same day. If something breaks, the whole company stops.

In a big-bang start people learn five new things at once. Mistakes pile up and nobody knows where they come from. Trust in the system drops in the first week.

The remedy is a staged implementation. Start with the process that hurts most, let it run for a month, and only then add the next one.

Many owners worry that stages will make the project longer. In practice the risk is lower, because any mistake affects only one process.

A good first stage is one people will feel straight away: the stock nobody can vouch for, or the orders that get lost in email.

Stages have another benefit. Each new process starts with people who already know the system, so the second stage moves faster than the first.

2. The software is chosen before the process is mapped

Many companies watch demos before they have written down how they work today. Then they choose by feature list and by the nicest screens.

After go-live it turns out an important step is missing. A specific costing, for example, or a manager’s approval. Workarounds in Excel begin and the system stays half empty.

Take one real order and follow it from enquiry to payment. Write down who does what and where it waits.

While you map the process, note three things:

  • Which documents come in and go out at each step.
  • Where data is retyped from one program into another.
  • Which exceptions happen every week, not once a year.

You don’t need a consultant for this. An afternoon, a sheet of paper and the people who do the work are enough.

The full sequence of steps is in the guide to choosing an ERP.

3. Nobody inside the company owns the project

The vendor can configure the system, but it can’t make decisions for you. Someone has to say what matters, when it’s done and what can wait.

Without that person, questions sit for weeks. Each department pulls in its own direction, and the owner hears about problems from unhappy staff.

  • Name one person who knows the process and is allowed to decide.
  • Give them real time for the project, not “whenever there’s a gap”.
  • Meet the vendor every week for a short review.

That person doesn’t have to be the owner. Often the best choice is an experienced team member from the warehouse or sales who knows every workaround.

4. Data is left for the last week

Items, customers, prices and opening stock all have to go into the new system. If they are spread across five files with different names, the transfer takes longer than expected.

Messy data brings mistakes from day one. One item under three names means three different stock levels, and none of them is right.

For example, a company with 900 items might find that 150 of them are duplicates or old codes. The numbers are hypothetical.

Decide who owns each master list: items, customers, prices. Without an owner, the clean-up stops halfway.

Run at least one trial import weeks before go-live, so problems surface in time.

Start cleaning while you are still choosing software. We describe the steps in the article from Excel to ERP.

5. People meet the system on go-live day

The people who will use the program every day often see the screens for the first time at go-live. Naturally, they go back to the old way at the first difficulty.

Getting people on board starts much earlier. Show them the system with their real tasks, ask what feels awkward, and fix it before go-live.

Your best ally is the person who loses the most time retyping today. They will see the benefit first and show it to their colleagues.

Classroom training rarely works. People remember what they did themselves, with their own order and their own item.

  • Show the system to key people at least two weeks before go-live.
  • Have everyone enter one of their real orders or deliveries as a test.
  • Prepare short “how to” sheets for the most common tasks.
  • Be on site during the first days after go-live, not just on the phone.

6. The cost grows because the scope isn’t written down

If the quote says “ERP implementation” with no detail, every new request becomes extra work. The budget grows and both sides run out of patience.

Ask for a written scope: which processes, which reports, which links to other programs, and what is left for a later stage. The price should cover that scope, with changes estimated separately and in advance.

Compare the cost over three years, not just at the start. What goes into it is explained in the guide to ERP cost.

Watch out for per-user pricing too. When every new hire means another licence, companies limit access and people go back to Excel.

Also ask what happens if you decide to change vendor. Your data should be exportable in a standard format.

Checklist: is your project on track

Go through the list with the person who owns the project. Every “no” is a task for this week, not for after go-live.

How we approach it

We implement in stages, starting with the process that hurts most. The project price is clear after the analysis and before we start. You talk directly to the people who write the code, so questions don’t sit in a queue. See how we work, step by step.

Frequently asked questions

How long does an ERP implementation take for a small company?

It depends on the scope and the data, so an honest answer comes after analysing the process. With a staged implementation, the first process works long before the last one.

What should I do if our ERP project is already going wrong?

Stop adding new things and pick one process to get fully working. Write down what doesn’t work, who owns it and which data needs fixing.

Is a large ERP with lots of features the safer choice?

Not necessarily. More features mean more configuration and more to learn; what matters is that the system follows your process and can grow with you.

Show us how you work today.

In 30 minutes we’ll find the process that costs you the most and see how it would look in Leanovix.

Book a demo