Choosing stock control software: the questions that actually separate them

Put four stock control systems side by side and the feature lists are indistinguishable. All of them do items, locations, stock levels, barcodes, reports and users. All of them demo beautifully, because a demo is somebody who knows the software driving clean data.

The differences are real, but they live in places a feature list does not go. Here is what to ask instead — including the questions that are uncomfortable, which are usually the informative ones.

A note on our position: we make one of these. Everything below is written to be useful whoever you end up choosing, and some of it does not flatter us.

1. What happens at the shelf?

The single biggest predictor of whether a stock system works is how long it takes to record taking something. Not how good the reports are — nobody has good reports on bad data.

So ask to see the taking-something-out flow, on the device it will actually run on, done by somebody who is not the salesperson. Count the taps. Anything over about fifteen seconds at the point of issue will decay into people not doing it, and then you have an expensive spreadsheet.

Ask them to demo it with gloves on. It sounds facetious. It is the actual operating condition, and it eliminates half the interfaces immediately.

2. Does it distinguish things that come back from things that get used up?

A box of washers and a torque wrench are different shapes of problem. One is a quantity that goes down; the other is a specific object with a holder, a due date and a history.

Plenty of systems model only the first and ask you to fake the second with a quantity of one. It works for about a month, until you need to know which of the four drills is overdue and from whom. If you have tools as well as consumables, this question narrows the field faster than any other.

3. Can I get my data out, today, without asking?

Not "is there an export" — every system says yes. Ask specifically:

• Can I export everything, including the transaction history, or only current balances? A system that exports balances but not movements is holding your history hostage.
• Can I do it myself from the interface, or does it require a support ticket and a fee?
• What format? CSV is fine. A proprietary backup you can only restore into their system is not an export.
• If I stop paying, what happens? Can I still read my data, and for how long?

That last one deserves a written answer. It is the question that reveals how a supplier thinks about the relationship, and the answers vary enormously.

4. What does it actually cost at the size I will be in two years?

Per-seat pricing deserves particular scrutiny in a stores context, because it gives you a financial reason to share logins — and a shared login destroys the accountability the system exists to provide. If a supplier charges per person and your store is used by forty people, ask directly how they expect that to work.

5. How does it handle being wrong?

Every stock system will hold an incorrect figure at some point. What distinguishes them is what happens next.

1. Can somebody edit a balance directly? If yes, ask who can, and whether the previous value survives. A system where a balance can be overwritten without trace cannot be audited, and the correction is indistinguishable from a cover-up.
2. Is a stock count recorded as an adjustment with a reason and a person, or does it just overwrite?
3. Can you see the history of one item — every movement, with who and when — on one screen? This is the screen you will actually use when something does not add up.

6. What does support look like on a Tuesday?

Ask for the response time in writing, ask whether it is a person or a portal, and ask what happens out of hours. Then ask the question that gets the honest answer: what is the most common thing customers contact you about? A supplier who says "nothing really" is either not listening or not telling you.

The questions worth less than they seem

• The number of integrations. What matters is whether it integrates with the one or two things you actually use.
• Mobile app versus web. What matters is whether it works on the device at the shelf, not what technology it is.
• AI features. Ask what decision it makes and what happens when it is wrong. Usually the answer is a forecast you would not act on unprompted.
• The size of the customer list. A system built for a thousand-line warehouse is not necessarily good at a two-hundred-line maintenance store, and often the reverse.

For what it is worth, Nilstock is priced per site rather than per person specifically because of the shared-login problem above, treats consumables and equipment as genuinely different things, and exports everything including the movement history without asking. The pricing page has the numbers and the feature list the detail. If you are still on a spreadsheet, when a stores spreadsheet stops working is the more useful place to start; if you have decided already, the first thirty days is the plan.