Think Build Implement Repeat
London, UK +44 7367 067226
WhatsApp FOLLOW f in X
  1. Home
  2. Blog
  3. We Only Find Out Stock Has Gone Missing at the Annual Stocktake. Can We Spot It Sooner?
Problems We Solve

We Only Find Out Stock Has Gone Missing at the Annual Stocktake. Can We Spot It Sooner?

Shrinkage discovered months later cannot be traced. We use machine learning on your stock movements to flag unusual losses by site, product and pattern early.

Updated 3 min readBy SpiderHunts Technologies

Free estimateNo obligation

Get a free estimate

Tell us what you need. A senior engineer reads every enquiry.

Takes under a minute. We never share your details.

  • Free consultation
  • No commitment
  • NDA on request

Prefer to talk? Book a free 30-minute call →

Quick answer — TL;DR

Shrinkage found at the annual stocktake is too late to trace: the pattern is long gone. Anomaly detection on everyday stock movements, adjustments, returns, write-offs and till data can flag unusual losses by site, product, shift or process within days, so someone can look while the trail is still warm.

The number at the bottom of the stocktake

The stocktake finishes and the variance is bigger than last year. Stock is missing, but nobody can say where it went. Theft, damage, mis-picks, receiving errors, unrecorded write-offs, a supplier short-delivering: any of them could explain it, and after a year the trail has gone cold. The loss gets written off and everyone hopes next year is better.

In between stocktakes, the system shows the stock as present. Orders are promised against it. Then a picker finds an empty location, a customer is let down, and someone does a quick adjustment to make the numbers match.

Why shrinkage stays hidden

Stock systems record movements but rarely question them. An adjustment, a write-off, a return to stock or a receipt short of the delivery note all go through without anyone asking whether it looks normal. Across thousands of movements a week, a small, steady loss on certain lines or at certain sites is invisible.

Cycle counts help, but they are usually scheduled by location or rota rather than risk. The counters check shelves that are fine and miss the ones where the problem is. And when a count does find a gap, it is corrected rather than investigated.

The adjustment itself then hides the evidence. Once someone has corrected the system to match the shelf, the loss looks like a routine stock correction among hundreds of others. Unless something is looking across all those corrections for patterns, such as the same product, the same site or the same week of each month, the cause never surfaces.

What late discovery costs

Hidden lossWhat follows
Missing stockWrite-offs that cannot be traced to a cause
Phantom stock in the systemOrders promised against items that are not there
Process errors repeatedThe same receiving or picking mistake happens all year
Internal theft undetectedLosses grow and trust erodes when finally discovered
Blanket controlsEveryone is treated as suspect because the real source is unknown

Most shrinkage is process error rather than theft, and process errors are fixable, but only if you know which process, which site and which products.

How we flag unusual stock losses early

  1. We pull stock movement history from your stock system, WMS or ERP: receipts, picks, transfers, adjustments, write-offs, returns and counts, plus till data for retail sites.
  2. We build a picture of normal behaviour for each product, site and process, such as how often a line is adjusted, typical variance at a count, and normal return rates.
  3. We run anomaly detection that flags patterns out of line with that normal: repeated negative adjustments on one line, variances concentrated on certain shifts, receipts often short from one supplier, or high-value items with unusual void or refund patterns.
  4. Each flag comes with the evidence, so an operations manager can see why it was raised and decide whether to investigate.
  5. We point cycle counts at the flagged locations and products, so counting effort goes where the risk is.
  6. Investigation outcomes are recorded and fed back, so the system learns which flags matter and which are normal quirks of your operation.

We design the flags around processes, products and sites first. Where people are involved, flags are reviewed by managers with context, and treated as a reason to look, not as proof of anything.

What operations sees

A short weekly list of unusual loss patterns, each with the movements behind it. Cycle counts that go to the places most likely to have a problem. Causes found while they can still be fixed: a receiving check that is being skipped, a supplier who is routinely short, a product that gets damaged in one part of the warehouse.

The stocktake still happens, but it becomes a confirmation rather than a surprise.

And because causes are found, controls can be targeted. Instead of tightening rules for everyone after a bad stocktake, you fix the specific process or supplier that was losing stock, which staff find far fairer.

Is this your situation?

  • Stock losses are mainly discovered at stocktake.
  • Variances are written off without a known cause.
  • The system shows stock that is not physically there.
  • Cycle counts follow a rota rather than where problems are likely.
  • You have several sites or a large number of product lines.

FAQ

Frequently asked questions

The questions readers ask us after this guide.

Still have a question?

Ask us directly — a senior engineer will get back to you.

Ask about your project

Is this about catching staff stealing?

Mostly not. Most shrinkage comes from process errors. The flags point to where losses concentrate, and managers decide what to look at with context.

Do we need new hardware like RFID?

No. The analysis uses the movement data your stock system already records. Better data capture helps, but it is not required to start.

How soon after a loss would it be flagged?

That depends on how often data is refreshed and how clear the pattern is. Frequent refreshes and regular counts both help surface patterns sooner.

What drives the cost?

The number of sites and systems, the volume of movements and how clean the movement codes are.

What do you need from us?

Stock movement history, stocktake results, and till data for retail sites, plus someone who knows how each movement type is used.

Keep reading

More on Problems We Solve

Start here

Tell us what your data should be telling you

Describe the decision you want to improve and the data you already keep. We will give you a straight answer on whether machine learning fits, and if a report or a simple rule would do the job, we will say so.

  1. You tell us what you needTwo minutes on the form, or a message on WhatsApp.
  2. A senior engineer reviews itAnd comes back with questions, a realistic range and an honest view on fit.
  3. Free 30-minute scoping callWe talk through scope, options and a realistic estimate — with no obligation.
Free estimateNo obligation

Talk to someone who builds this

Send a short brief and we will come back with an honest view and a realistic range.

Takes under a minute. We never share your details.

  • Free consultation
  • No commitment
  • NDA on request

Prefer to talk? Book a free 30-minute call →