Process Before Platform

I've worked with decorated goods businesses that spent between £50,000 and £500,000 on new ERP systems. They spent 12 to 18 months implementing. They went live. And in most cases, nothing significantly improved. Sometimes it got worse.

The reason is always the same. They bought the platform before they understood the processes. The new software arrived on top of broken operations and digitised them. The mess just moved faster.

What Software Actually Does

Software automates and accelerates. That's its value. But it automates and accelerates whatever is already happening — including the inefficiencies, the workarounds, the unclear handovers, and the processes that depend on one person who knows how things actually work.

If you have a broken order-to-production process, a new ERP doesn't fix the broken process. It puts it into a more expensive system. The order still gets lost between sales and production — it just gets lost in a different place. The artwork still doesn't reach production with the order — it just doesn't reach production faster.

Software is a lever. What it amplifies depends entirely on what's underneath it.

How to Know If Your Processes Are Ready

The test is simple: can you describe your order-to-despatch process, step by step, and have three different people in the business independently describe the same process in the same way?

If the answer is no — if different people describe different steps, different handover points, different rules — then the process isn't clear. And if the process isn't clear to the people doing it, implementing it into a new system won't make it clear. It'll implement one version of it and create conflict with everyone who thought the process was different.

The process has to be agreed before the system is configured. Not during implementation. Not after go-live. Before.

Mapping Your Actual Process

The most useful thing to do before evaluating any new platform is map how work actually flows in your business — not how you think it flows, and not how the manual says it should flow.

The method is straightforward. Follow a single order from the moment it arrives to the moment it's despatched. Write down every step. Note every person involved. Note every tool or system used at each step. Note every decision point and who makes it. Note where the order waits — and why.

Then follow a problem order — one that went wrong or took longer than it should. Map the same things. The gap between the two maps describes your actual process: what works, where it breaks, and why.

Most businesses find, when they do this exercise properly, that they have three to five significant process failures that are costing them time and money every week. Most of those failures can be addressed without new software. Fixing them first makes the subsequent software implementation — if one is needed — significantly more likely to succeed.

The Three Bottleneck Rule

When you've mapped your processes, you'll typically find more problems than you can fix at once. The prioritisation framework I use is simple: identify the three bottlenecks that are costing you the most time, money, or customer satisfaction. Fix those first. Don't start on the rest until you've fixed the three.

The reason is resource and attention. Trying to fix ten things at once means nothing gets fixed properly. Fixing three things completely and verifiably gives you a better operation and momentum for the next three.

Once the top three are fixed and the processes are stable, the question of whether new software is needed becomes much clearer — because you now understand what the software needs to do, and you can assess whether what you have can support the improved process.

When You Actually Need New Software

You need new software when your processes are clear, documented, and consistently followed — and the system you have genuinely can't support those processes at the scale or in the way you need.

Not when the system is old. Not when a vendor tells you it's time to upgrade. Not because a competitor has a new system. When the process is clear and the current system is a genuine constraint.

At that point, buying new software makes sense — because you know what it needs to do, you can configure it correctly from the start, and you can measure whether it's actually doing what you bought it for.

Common Questions

Why do ERP implementations fail in decoration businesses?

The most common cause is implementing software onto broken processes. The new system automates what was already happening — including the inefficiencies. The implementation adds cost and disruption but doesn't fix the underlying problems because those are in the process, not the software.

How do you map your actual business processes?

Follow an order from arrival to despatch — physically if possible. Note every step, decision, person, and tool. Then follow a problem order. The gap between the two describes your real process: where it works and where it breaks.

When do you actually need new software?

When your processes are clear, documented, and consistently followed — and the current system is a genuine constraint. Not because it's old, not because a vendor says so. When the process is solid and the system can't keep up.