Real time inventory tracking means the stock figure changes when stock physically moves, not at the end of the shift, after a nightly sync or during the month-end count. Every receipt, putaway, transfer, pick and despatch is recorded as it happens, so the warehouse and every sales channel read the same number. When the record is current, you know what is on the shelf and your website can sell with confidence. When it lags, the figure on screen becomes a guess.
Most of the damage comes from small gaps. A sale on Amazon, eBay eller Shopify can take stock that the warehouse has already sent out, and the first sign of trouble is a cancelled order. This guide covers what real time means on the floor, what breaks it, why overselling is usually a timing problem, how to keep one view across several sites, where scanning stops being enough and what to ask a vendor before you buy.
What real time inventory tracking means in a warehouse
Real time inventory tracking records each stock movement at the point it happens, through a scan or a system transaction, so the stock record matches the shelf within moments. The record changes when the pallet moves, not when someone gets round to typing it in.
In practice, that means treating every movement as a transaction. A pallet is received, put away to a location, transferred, allocated to an order, picked, packed and despatched. Each step either scans something or triggers a task that a scan completes. Mathew Buttar, Head of Solutions at Clarus, put the effect this way in an internal interview: if the supplier sends the load data up front, a truck with 26 pallets needs 26 scans, and the whole load is received. The scan does the work that a stock adjustment used to do.
| Stock event | Real time setup | Batch or spreadsheet setup |
|---|---|---|
| Goods received | Pallet scanned at the dock, stock record created | Goods listed on a paper note, keyed in later |
| Put away | Location and pallet scanned, stock sits in that location | Location written down, recorded at end of shift |
| Stock transferred | Scan moves the stock and both locations update together | Adjustment entered after the move, if at all |
| Order picked | Pick confirmed by scan, stock leaves the available pool | Pick list ticked off, stock deducted in a nightly run |
| Despatched | Dispatch scan closes the order and can send a file onward | Shipment reconciled against the invoice at month end |
The table makes one point. In a real time setup, the stock record moves with the warehouse. In a batch or spreadsheet setup, it moves after the warehouse, and the gap between the two is where errors live.
The same logic runs through the rest of the process. Mathew Buttar also described a pallet that is received with one scan, transferred with one or two, picked with one and dispatched with one, with a file sent on to the next business in the chain. Every hand-off is a record, not a reconciliation exercise. For the scanning side of this, see the Clarus inventory management feature page.
What breaks real time inventory tracking?
Real time tracking usually fails in one of three places: the nightly batch sync, the spreadsheet that runs beside the system, and the posting that happens days after the movement. Each looks harmless on its own. Together they produce a figure that is wrong often enough that nobody trusts it.
Batch syncs run on the wrong clock
A batch sync that runs overnight means the figure is up to a day old. Picks and sales made in the morning draw on yesterday’s stock, and a short-dated line can be picked from a location that was emptied at lunchtime. Batch runs also hide problems. An error made at 9am sits unseen until the next run, and by then more orders have been built on top of it.
Spreadsheets hold the truth for one person
When the stock count lives in a spreadsheet, its accuracy depends on whoever last updated it. There is usually no audit trail and no clear answer to who moved what. Paper pick lists create the same problem at the shelf. Mathew Buttar’s point, drawn from his earlier warehouse work, is that a paper pick list needs a second person to walk round the pallet and check it against the list, so one correct pick takes two people’s labour. A scan removes that check, because the system knows what was scanned. The cost of a spreadsheet shows up most in a recall. Food and product businesses audited against BRCGS standards need traceable records, and a figure that lags the shelf makes a recall slower to run.
Delayed postings and goods in backlogs
Stock arrives without warning more often than most plans assume. Mathew Buttar says unannounced stock happens all the time, and that a simple yes to a customer can become a green light for them to send whatever they like. When goods in falls behind, picking falls behind too, because the pallets that need to move to make room are not moved in time. The receipts waiting in the yard cannot be posted, and the figure drifts further from the floor. If you also run counts and adjustments, our guide to inventory control software covers that side of the job.
Why does overselling happen, and why is it usually a timing problem?
Overselling rarely starts with a miscounted shelf. It starts with a gap in time. A sale on one channel takes stock that has not yet been removed from the figure another channel reads. Each channel refreshes on its own schedule and reserves stock in its own way, so a slow channel can keep selling units that have already gone.
The practical fixes are ones any seller can apply. Hold a buffer back from each channel, sync stock as close to live as the integration allows, and keep one central record that every channel reads from. The buffer is a judgement call. Hold back too much and you turn away sales you could have made. Hold back too little and a busy week still produces cancellations.
Clarus connects to the main marketplaces and carriers through its integrations library, and orders arrive in the same system that allocates stock to them. That matters because the order and the stock it needs sit in one place. If you sell online only, our guide to programvara för lagerhantering för e-handel looks at the channel side in more detail.
How do you keep one view across several sites?
Real time visibility gets harder with every extra site, because each site adds another record that can drift. The answer is one stock record across all sites, with locations that belong to a site and movements that update the same figure wherever they happen. Clarus runs multi-warehouse and multi-site operations from a single environment, so a stock take at one site and a dispatch at another sit in the same record.
Visibility also matters outside the building. A 3PL’s client wants to know what stock they hold without phoning the warehouse. DGS Transports deployed live warehouse visibility and a client portal to replace a legacy system. A portal is only as good as the record behind it, so it tests the tracking as much as the presentation.
Mathew Buttar has also said that a new warehouse can be set up in minutes if the location and stock data is ready. He adds two caveats: getting scanners running at a new site takes longer, and the speed only holds when the data is in place. Plan for both.
Where does scanning stop being enough?
A scan confirms a barcode. It does not confirm what happens to the item next. Mathew Buttar’s description of the failure is blunt. An operator can scan the location, scan the right pallet and scan the right product, then take a box from the pallet next to it. Every scan was correct and the pick was still wrong.
Clarus narrows that gap in a few ways. Streckkodsavläsning can be set up so that a scan is rejected when the barcode does not match the order line. Handheld workflows can be built around each site’s process, so the operator is prompted for exactly the data the stock record needs. The hardware is a separate choice, and handheld makers such as Zebra build the devices that run these workflows.
Be clear about the limit, too. The live transaction grids in Clarus show what happened, who did it and when, but they are built from actions people took in the system. They cannot see a box leaving a pallet. Damaged pallets are a good test of this. A pallet damaged on receipt should go on hold, not into racking, and workplace safety in UK warehouses, racking included, is overseen by the Health and Safety Executive. Mathew Buttar’s summary is the one to keep in mind: there is not a warehouse in the world that does not have errors, so the job is to reduce them and catch them early.
What should you ask a vendor before you buy?
Most vendors will say they offer real time stock. Ask them to show it. These questions separate a live record from a report that is simply refreshed often.
- When does the stock figure change? Ask whether it changes at the scan, on a batch run or when someone posts the movement. Ask them to receive, put away and pick a pallet in the demo and watch the figure move.
- Can a movement skip the scan? Ask where an override is allowed and whether it is logged with the user and the time.
- How do sales channels read stock? Ask for the refresh interval in writing, and whether a channel sees live stock or a snapshot taken at intervals.
- What happens when a sync fails? Ask how a failed file or API call is detected, who is told, and whether the stock figure is held until the update is sent again.
- Can stock be held automatically? Ask whether received stock can go straight to a quarantine location and stay out of allocation until someone releases it.
- Can you follow one pallet? Ask for the full history of a single pallet, from goods in to despatch, pulled up live in the demo.
- Which reports run on live data? Ask which dashboards read the live record and which run on scheduled extracts. Our guide to bespoke WMS dashboards covers how reports help teams find and filter stock faster.
Stock accuracy is the number to track alongside all of this. Our warehouse KPI guide sets out the measures worth reviewing each week.
What does good look like in practice?
JODA Frakt reached 99% stock accuracy by year two, having moved from stock takes that ran for weeks on Excel to a count that takes a day. realtidstrategispel also reports 99% stock accuracy across Europe on a cloud WMS. Before you accept any accuracy figure, ask how it was measured. A number is only useful if you know how it was counted.
The verdict: real time inventory tracking pays off when the scan is the record. If your stock figure is updated by a batch run, a spreadsheet or someone typing it in later, you have a reporting tool, not a live view. Start by mapping where each movement is recorded, then close the gaps that your sales channels feel first.
Tala med en lagerarbetare
Om du håller på att utvärdera dina alternativ och vill se hur ett specialutvecklat lagerhanteringssystem (WMS) fungerar i praktiken, är Clarus värt att ta en pratstund med. Vi samarbetar med 3PL-företag och distributörer över hela Storbritannien för att implementera lagerhanteringsprogramvara som anpassas efter just er verksamhet – inte tvärtom.
Kom i kontakt med vårt team att diskutera dina krav.
Related reading: how to improve inventory accuracy.