The inventory count sheet: what goes on it, and what stays off

Most count sheets are a product export with one column added. The sort is alphabetical, the expected quantity sits in plain view, and the counter walks the room three times. The sheet's job is to make the count fast and honest, and both come from the layout.

This page covers the columns a count sheet needs, why the expected quantity stays off it, how to lay it out so a counter walks the room once, and what happens to the sheet afterwards. The sheet is the same whether the count is the annual full count or one of the rolling counts in cycle counting cadence.

What columns does an inventory count sheet need?

Seven: location or bin, SKU, product description, unit of count, counted quantity, counter initials, and time. Add a recount column for the referee. Everything else, the expected quantity, the cost, the supplier, belongs to the reconciliation afterwards, not to the floor.

The unit column matters more than it looks. The same product can sit on the shelf as single units, as inner packs and as the full cases a stockist orders. If one counter writes 12 for a row of cases and the next writes the number of units inside them, the sheet reports a variance that does not exist. Print the unit the count is made in, count in that unit, and convert once at the desk.

The description is there for the counter, not the system. Two flavours in the same carton, two pack sizes of one product, a case that was relabelled: a SKU code on its own does not tell a counter which is which, and a counter who cannot tell them apart counts them together. Write the description the way the floor names the product.

Why should the expected quantity stay off the sheet?

Because a counter who sees 48 tends to find 48. A blind count removes the anchor, so a variance is a measurement rather than a confirmation. The comparison happens at the desk, where the counted quantity is matched to the system by SKU and location.

The objection is always speed: with the expected number in view, a counter can skip the lines that look right. Those are exactly the lines that should not be skipped. The cost of a blind sheet is a slower first count; the cost of an anchored one is a variance list that looks clean and is not, which is how the same shortage survives three counts in a row.

The one exception is a recount. The referee may see both numbers, because the question has changed from how many are there to which of these two counts is right. Keep the recount in its own column or on its own sheet, so the first count is never written over.

How should the sheet be laid out?

In walking order. Sort the lines in the sequence a counter meets them: aisle, bay, shelf. One sheet per zone, the zone name at the top, page numbers, and a sign-off line for the pair at the bottom. A sheet sorted by SKU sends the counter back and forth across the room.

Walking order is a property of the location, so a sheet exported from the product list has to be re-sorted by bin before it is printed. If stock is not binned, the count is the moment to bin it: a zone name and a shelf label written on the sheet today become the location column next time. A brand that holds stock in two places prints one set per site and never lets one sheet cross a wall.

Leave a notes column for anything the counter sees that the sheet did not expect: stock in the wrong bin, a damaged case, a product with no label, an opened case pack. Those notes are often the explanation for a variance found an hour later at the desk, and they are lost if there is nowhere to write them.

A spreadsheet printed per zone is enough for an annual count and for monthly cycle counts up to a few hundred lines. Paper stops working when counts happen weekly across more than one location and keying the sheets in takes longer than the counting did. Until then, the printed sheet has one advantage no screen has: the expected quantity is not one tap away.

What happens to the sheet after the count?

It is keyed or scanned against the record the same day, and kept. Every variance gets its sheet line as evidence: who counted, when, which zone. Keep sheets for at least a year, because next year's recurring variances are found by reading this year's sheets.

Match by SKU and location, not by SKU alone. A product that is right in total and wrong per location is the most common pattern in a brand that ships from two places, and the total hides it. Sort what is left by value, not by line count, and name a cause for each before any quantity is adjusted; most of the value usually falls into the four causes in inventory shrinkage.

The sheets from this count are the starting data for the next one. Lines that drifted go on the frequent list in cycle counting cadence; zones that came back clean can be counted less often. Without the sheets, the next count starts from zero again.