Stock Counts and Shrinkage: Keeping Track

Why this matters

Shrinkage, the polite word for stock that disappears without being sold, eats 1–3% of revenue in a typical retail business. On a shop doing £500,000 a year, that’s £5,000–£15,000 walking out of the door through shoplifting, staff theft, admin errors, and damaged goods nobody recorded. Most independent retailers have no idea what their shrinkage rate is, because they’ve never properly counted. You can’t reduce a number you don’t know. Stock counts feel like a chore; they’re actually a profit-recovery exercise.

Practical steps

1. Count stock properly, and regularly

The annual stocktake is not enough. By the time you count once a year, a year’s worth of problems have compounded beyond diagnosis:

• Full counts at least twice a year, with the shop closed or during quiet hours. Everything counted, everything recorded, variances investigated, not just noted.
• Cycle counts in between: count a different category each week on rotation. High-value and high-theft categories (small electronics, premium cosmetics, anything pocket-sized and expensive) get counted most often. A rolling cycle means every line gets counted several times a year, and discrepancies surface within weeks, not months.
• Count blind where possible, the counter doesn’t see the expected figure first. Expected figures create “close enough” counting.
• Record who counted what. Accountability improves accuracy; anonymous counts don’t.

2. Understand what the numbers mean

Shrinkage is calculated as: (expected stock value − actual counted value) ÷ sales, expressed as a percentage. But the headline number is just the start, the breakdown tells you where to act:

• Known losses (recorded damage, documented markdowns, legitimate disposals): these should already be in your records. If they’re not, your admin is the problem.
• Unknown losses (the gap after known losses are accounted for): this is theft, unrecorded damage, and receiving errors. This is the number to drive down.
• Category-level shrinkage: if one category loses 8% and the rest lose 0.5%, you don’t have a shop problem, you have a category problem. Lock it up, move it, tag it, or watch it.

3. Investigate shrinkage like it matters

When a count reveals a significant variance, investigate it as a process, not a witch hunt:

• Check receiving records first, was the stock ever actually delivered? Supplier short-deliveries miscoded as shrinkage are common and embarrassing.
• Check the sales records, were items sold but rung up wrong, or discounted without authorisation?
• Check damage and disposal logs, was it written off but never recorded?
• Only after ruling out admin errors do you look at theft, internal or external. And if it is theft, your incident log and CCTV records are what make the case, which is why those records need to exist before you need them.
• Document the investigation and its conclusion. “Counted 12 short, found 8 were a receiving error on 14th, 4 unexplained, moved to locked display” is a closed loop. Unexplained variances with no follow-up are just numbers.

4. Act on the patterns

Counting without action is just expensive arithmetic:

• High-theft lines: move to locked cabinets, add security tags, reposition near the till, or reduce facings.
• Receiving errors: tighten goods-in checks with a delivery log signed at receipt.
• Staff-related patterns: review cash-handling and discount authorisation procedures, and remember that most staff theft is opportunistic, enabled by weak processes. Fix the process.
• Review shrinkage figures monthly against the cycle counts. The trend matters more than any single count.

Common mistakes

1. Counting once a year and calling it done. Annual counts tell you how much you lost, not when, where, or how. By the time you find the problem it’s a year old. Cycle counts catch problems while they’re fixable.
2. Not recording known losses. Damaged goods binned without a record, markdowns without paperwork, samples given away, all of this looks like theft in the numbers. Record every legitimate loss and your “unknown” shrinkage often halves overnight.
3. Treating every variance as theft. Most shrinkage is process failure: receiving errors, pricing mistakes, unrecorded damage. Accusing staff before checking the paperwork destroys trust and usually misses the real cause.
4. Counting without acting. The stocktake reveals the same category losing 6% for the third year running, and nothing changes. A count that doesn’t lead to a locked cabinet, a moved display, or a tightened process was a waste of a Sunday.

Templates that help

• Retail Stock Count Sheet Template, structured count sheets for full stocktakes and cycle counts, with variance fields.
• Retail Loss Prevention Checklist Template, routine security checks to run between counts.
• Retail Incident Report Template, record shoplifting and suspicious incidents to correlate with count variances.
• Retail Supplier Order Form Template, delivery confirmations that prove what was actually received.
• Retail Daily Sales Log Template, sales data to set against stock movements when investigating variances.
• Retail Return Exchange Form Template, returns are stock movements too; unlogged returns look exactly like shrinkage.

Bottom line

Shrinkage is a number you can only manage if you measure it: cycle-count high-risk categories weekly, investigate variances as process failures before assuming theft, record every legitimate loss so the “unknown” figure is real, and act on what the patterns tell you. The shops with the lowest shrinkage aren’t the ones with the most cameras, they’re the ones that count, investigate, and follow through.