What is an ERP, and when does your company actually need one?

The real problem an ERP solves (it is not “organisation”)

An ERP does not tidy your company: it stops every area from working with its own version of the truth.

Without a central system, each area decides with incomplete or stale information. It is not a discipline problem: nobody is looking at the same number at the same time.

These are the symptoms that show up in every industry:

• Sales promises a delivery date without knowing whether the stock actually exists.
• Finance closes the month with numbers that do not reconcile because each area reported separately.
• Purchasing duplicates orders because it cannot see what another site or channel already ordered.
• Management decides on reports that are two or three weeks out of date.

The signs you need one now, not in two years

With two or more of these signs, operating without a system already costs more than implementing one.

You rebuild the reports by hand every month, combining three or four sources in a spreadsheet. Nobody has a complete view of inventory, sales and finance at once. Double data entry already produces losses or customer complaints.

Scaling means hiring more people to do the same work manually, instead of the system absorbing the growth. And every area uses its own tool, with none of them talking to each other.

If you recognised yourself in two or more, the cost of carrying on has already passed the cost of fixing it.

Generic ERP vs. custom system: the decision nobody explains well

A generic ERP covers 80% of any company; the remaining 20% is what gets expensive.

A generic ERP is designed for the average company's processes. The catch is that almost every business has its own — that remaining 20% — which ends up forcing costly adaptations, modules nobody uses, and per-user licences that climb as the team grows.

The right question is not “which ERP is best?” but “is my process standard enough to pay for a system built for any company?”

The modules an ERP usually covers

Almost no company needs all six modules on day one.

A full ERP typically ships inventory and warehousing (real-time stock, multiple locations, traceability), sales and invoicing (quotes, orders, e-invoicing), purchasing (orders, suppliers, cost control), finance and accounting (cash flow, receivables and payables, reporting), human resources (payroll, shifts) and, where relevant, production.

One of the most expensive mistakes is implementing all six when the real pain is in two or three. Every module switched on without need adds licences, training and maintenance surface.

How to work out the real cost, not the licence price

The number that matters is total cost over three years, not the first monthly fee.

An ERP's list price is the visible part. Underneath sit the extra users you will need, the modules bought later, the implementation consulting and the weeks of productivity lost while the team learns the tool.

Do the full exercise before signing: project licences, implementation and maintenance over 36 months, and compare that against building only what you actually use. The figure often changes the conversation entirely.

How to decide without overspending

Start from the concrete pain, not from the software category.

Identify the real problem first, not the product: is it inventory? Is it that finance does not reconcile with sales? Start there, not from “we need an ERP” in the abstract.

Assess how standard your process really is. If your business runs like any other in your industry, a well-implemented generic ERP can be the fastest route. If your competitive edge is precisely that you operate differently, forcing yourself into a generic mould can cost you that edge.

And consider starting with what is critical: a system focused on the 20% of the operation that hurts, rather than a full ERP that will take months to go live.