Inventory rarely goes wrong because nobody knows what is in stock. It goes wrong because nobody knows what is still free. This page covers the number an inventory management system is built around, how reorder points should actually be calculated, where a spreadsheet stops working, and when inventory software buys you nothing.
By Tom Joseph · Last updated: 18 August 2026
An inventory management system maintains four numbers per item: what is physically there, what is already committed to orders, what is inbound from suppliers, and what remains available to sell. That last number — available to promise — is what the system exists for, and it is precisely the one a spreadsheet cannot hold reliably once several people or channels sell at the same time. We start with a €2,500 audit that establishes whether your problem is stock levels or stock locations, because those are different systems and different investments.
What the system decides for you
Not what is on the shelf, but what you can still promise from it — and where that number stops being trustworthy.

What sits there is not what you can sell. Reserved, blocked and already promised come off first.

Most reorder points were set once and never revisited. Demand moved; the number did not.

A spreadsheet works until two people edit it at once, or until nobody knows which version applies.

If you sell across several channels, overselling is a synchronisation problem — not a stock problem.

The cost nobody books. Stock that does not move consumes space, cash and attention.
Ask how many of an item are in stock and you get a physical count. That number is almost never the one a decision may be made on.
What you actually need is available to promise: physical stock, minus what is committed to open orders, plus what is inbound on a confirmed purchase order. Of 400 physical units, 380 may be committed — so you cannot sell another 100, however reassuring the shelf looks.
This is the dividing line between stock recording and inventory management. The first tells you what is there. The second tells you what you may promise, and that is the only number sales can use.
Almost every system supports a minimum and maximum per item. The problem is not the feature but where the numbers came from: set by hand once, against the demand of that moment, and never touched since.
A reorder point that works has three parts:
A spreadsheet is a perfectly reasonable inventory system for one person with a few hundred items and one sales channel. Starting there is not naive; staying there once one of these three things happens is.
If you sell through a webshop, a marketplace and a rep with an order pad, the question is not how much you have but how quickly each channel is told when it changes.
What decides this in practice:
Stockouts get attention because a customer complains. The opposite failure is silent: stock bought once, never sold through, still sitting there tying up cash and space while looking like an asset on the balance sheet.
Most businesses cannot say what share of their stock has not moved in a year, which is why a good system earns its cost here rather than on reordering:
These two get conflated, and the distinction decides which quote you request.
An inventory management system answers how much you have and what to reorder. It can work perfectly without you knowing a single warehouse location — plenty of webshops and wholesalers run on one for years.
A WMS answers where something is and who picks it next. You need that when finding and directing work is the constraint, not the arithmetic.
Every package books stock in and out. The differences that touch your operation are narrower:
If one person buys and sells, you have one channel, and the goods do not perish, a spreadsheet with discipline is cheaper and faster than any package. That is not an interim measure; for many businesses it is the end state.
The threshold is not revenue but the number of places where something happens to the same stock at once. One place: a spreadsheet is fine. Three or more: anything that does not hold the counts centrally will cost you money in stock you could not sell or sold twice.
If you are unsure, count first. If you do not know how far your current numbers sit from reality, you cannot justify any investment — and that count costs a day, not a project.
An inventory system manages quantities and reordering; a WMS manages locations and work.
Entry-level packages are priced per user per month; the costs that surprise people are the integrations.
Yes, as long as there is one place where stock changes. Three signals mean it is no longer viable.
Physical stock minus what is committed to open orders, plus what is confirmed inbound. It is the only number sales may promise against.
Calculate them from three parts rather than setting them by hand once.
The €2,500 audit establishes whether you need an inventory management system or a WMS — and says plainly when a spreadsheet with discipline is enough.
Book a conversation