Shrinkage

How shrinkage is worked out

The sum is the value of the stock that has gone missing, divided by the value of what you sold at cost over the same period, multiplied by one hundred. Some businesses divide by average stock value instead, which gives a bigger-looking number from the same loss. Either denominator works; moving between them from one year to the next does not.

The part worth more than the arithmetic is the split. Shrinkage is not one thing, and reported as a single percentage it is almost useless. Break it into the categories you can actually investigate:

• Goods damaged in the warehouse and never written off properly.
• Receiving differences: you were short-delivered, somebody signed the note, and the claim window closed.
• Picking errors shipped out and never charged back.
• Expiry and obsolescence on dated goods.
• Plain administrative error: a transaction booked twice, a unit of measure confused, a return received against the wrong line.
• Theft.

In most warehouses the first five together are larger than the sixth, and a count that starts from the assumption of theft makes the team defensive and the data worse. Count first, split second, draw conclusions last.

There is no good number

A jeweller, a builders' merchant and a frozen food distributor lose stock in entirely different ways and at entirely different rates, so an external figure is not a target. Nor is zero: a warehouse with no recorded shrinkage is usually a warehouse that is not counting, and the loss is sitting in the records waiting to be found at year end.

Compare against your own previous year, and by product group and by zone. A single zone or a single category well above the site average is a specific and findable problem.

In practice

The £24,000 difference on £2.4 million of goods sold at cost gives shrinkage of one per cent.

Investigated, it splits like this: £9,000 of damage that happened in the building and was never written off, £6,000 of receiving differences on three suppliers, £4,000 of expiry on dated lines, and £5,000 that nobody could explain at all.

Three quarters of the loss has a known cause and a known fix: a damage-recording habit, a counting discipline at goods-in, and tighter rotation on dated stock. The genuine mystery is £5,000, and it would have been invisible inside the one per cent. The figures here are an illustration.

What moves the figure

• Counting little and often. Regular counts on a rolling basis find a loss while the cause is still traceable. A single annual count finds a number nobody can explain. There is a cycle count calculator on this site.
• Counting at goods-in. Signing a delivery note is not receiving. Most receiving differences are found weeks later, after the claim window has closed.
• Making damage easy to record. If reporting a crushed carton takes five minutes and a supervisor, it will not be reported, and it becomes shrinkage.
• Location discipline. Stock in the wrong location reads as missing until somebody trips over it. That part of the figure is a stock accuracy problem wearing a shrinkage costume.
• Rotation on dated goods. First expiry, first out, enforced by the system rather than by memory.
• Access control, in proportion. Worth having, worth reviewing, and not worth treating as the whole explanation until the other five categories have been ruled out.