Thirty-five years of watching people count stock.

Thirty-five years, four times over

The job has never changed: get an accurate count off a shelf and into a system, without a person retyping it. What has changed, roughly once a decade, is the machine you do it with — and each time, the software had to be rebuilt from the ground up rather than ported.

1991–2008 — batch mode, and a cabinet full of it

The Psion Organiser first, doing real stocktakes in Australian warehouses while most people still thought of it as a pocket diary. Then the Nippon Denso BHT-3000, the Symbol PDT3100 series, and later the Symbol LDT-3805. Monochrome screens, a wand or a laser, a few hundred kilobytes of memory, and a serial cradle.

All of it was batch. Operators counted all day, docked the terminal, and the data went up that evening. A revolution over a clipboard — but you found out about a mistake the next morning, by which time the pallet had moved.

Every one of those devices is still in a cabinet here. None of them run any more.

2008 — ScanTrack Systems, and the end of batch

Windows Mobile arrived with colour screens, real databases and, crucially, wireless. That was the moment the count could stop waiting until evening, so we built ScanTrack Systems: an online portal with handhelds talking to it, covering goods in, putaway, moves, goods out and stocktakes. The programming was slow and tedious by today's standards, but it worked, and it is the same idea this product still runs on: the count is not yours until it has left the device.

Stocktakes Online came out of ScanTrack. Plenty of businesses wanted the counting and none of the rest, and asking them to carry a whole stock system in order to count once a year was the wrong deal. So the stocktaking was split out and sold on its own — which is what you are looking at, and part of why it is priced per stocktake rather than per year.

We used a marketing company that year, and the fellow gave us a slogan: Data Mobility for Business Agility. It was very 2008, and we used it for years. It described the shift honestly enough — the count had stopped waiting until evening — even if nobody has ever said “business agility” while standing in a warehouse holding a scanner.

2016 — the warning we did not act on

The badge on the devices had changed twice by then without much changing underneath: Symbol was bought by Motorola in 2007, and Zebra bought that enterprise business from Motorola Solutions in 2014 — largely the same people, still building the same scanners.

At a conference that year they said it plainly. Apple was never going to suit industrial. Windows Mobile was finished, with end of life set for 2020. Android was the future, and we should move. They were right on all three, and we should have moved then.

We didn't. Rewriting the whole thing for Android was expensive and would have taken months, and two things happened in the meantime that had nothing to do with barcodes: I got married in 2019 and our son was born in 2020. So instead of rebuilding, we kept the Windows Mobile fleet alive — more than fifty rental devices — and ran it right through to the end of 2024.

It worked for longer than it had any right to. But the hardware was old, spares got expensive, repairs stopped being worth it, and clients who needed a modern answer went elsewhere because we did not have one. That is the honest version. Being eight years late to Android cost us customers, and there is no version of this story where that was somebody else's fault.

2025 — a year with nothing to sell

The Windows Mobile product stopped at the end of 2024 and the Android one did not exist yet, so for a year there was no Stocktakes Online. People who asked were told that, and went elsewhere. It is the plainest bill for the decision not taken in 2016, and it is why the honest claim is that the idea started in 2008, inside ScanTrack, rather than that this product has run without a break ever since.

2026 — Android, wifi and the cloud

Warehouses have wifi now, and that changes the deal completely: there is no reason for the count to live on the device at all. Stocktakes Online sends every count to the portal as it is made, so a supervisor watching a screen sees the warehouse being counted in front of them, and a device dropped in a puddle takes nothing with it.

The rebuild that was out of reach for years is done. Thirty-five years of counting other people's stock is in it — which is why it argues with you about things that look like details: whether a repeated scan is a second pile or a correction, whether an empty bin is the same as one nobody visited. Those questions come from warehouses, not from a whiteboard.

Every generation of this software was faster than the last. Only this one removed the gap between counting and knowing.

What we actually do

We are not a software company that also sells hardware, nor a hardware reseller that bundles an app. We do the whole job: the scanners, the labels on the racking, the software that reads them, and the phone call when something is wrong on the morning of a count.

• Stocktaking software — Stocktakes Online, built for Zebra Android devices.
• Zebra hardware — supplied, staged and supported.
• Labels and printing — bin locations, magnetic rack labels, product labels, printers and media.
• Setting it up — location schemes, data loading, and being there for the first count.

How we work

Most of our customers are Australian businesses with somewhere between one warehouse and a handful of sites. They have no IT department, and they are not looking for a project. They want to count their stock accurately over a weekend and get back to work on Monday.

So we keep it plain: tell us what you are counting and when, and we will tell you what it costs and what you need. If we are not the right answer for you, we will say so — a stocktake that goes badly is not a sale we want.