Opening and Closing Stock Counts: How to Do Them Properly

Most small shops that lose track of stock aren't skipping counts altogether — they're doing them loosely enough that the numbers stop meaning anything. Someone counts the drinks fridge around midday, someone else eyeballs the shelf before locking up, and neither count is tied to a specific person or a specific moment. When a shortage turns up weeks later, there's no way to say when it happened or who was on shift, so nothing gets fixed and it just happens again.

Why the count has to happen at the actual shift boundary

An opening count is only useful if it's taken before a single sale happens that day, and a closing count is only useful if it's taken after the last sale. If your "opening" count actually happens two hours into trading, you've already lost the ability to tell whether a discrepancy happened overnight, during a delivery, or during those first two hours of business. The count is a snapshot — and a snapshot taken at the wrong time captures the wrong thing.

The same logic applies at closing. If the attendant locks up first and counts "whenever" the next morning, you've merged two shifts' worth of activity into one number, and you'll never be able to tell which shift the gap belongs to. The fix is simple in principle and hard in practice: the count has to be the literal first and last thing that happens on that shift, not a task squeezed in whenever it's convenient.

What to record — beyond just the number

A quantity on its own is close to useless if you ever need to investigate it later. At minimum, every count should capture:

• The quantity per product or category, not just a total value — a total can hide which specific item is actually short.
• Who did the count. Not "the morning person" — a name, tied to whoever was actually clocked in.
• The exact time the count was taken, so you can tell whether it happened at the true shift boundary or three hours late.
• A photo of the shelf or stockroom at the moment of counting, if you can manage it. A number can be miscounted or altered after the fact; a photo is much harder to argue with.
• Any stock that moved in or out during the shift — deliveries received, returns, damaged items pulled — logged separately from the count itself.

That last point matters more than owners usually expect. If a delivery arrived mid-shift and got added to the shelf, your closing count will naturally be higher than a simple "opening count minus sales" calculation would predict. If that delivery isn't logged, it looks like an unexplained surplus — and a surplus that isn't understood is just as much a sign of a broken process as a shortage is.

Reconciling an opening count against yesterday's closing count

This is the step most shops skip entirely, and it's the one that actually catches problems early. Yesterday's closing count should equal today's opening count, assuming nothing moved overnight. If it doesn't, you have a real question to answer before the day even starts — not three weeks from now when a monthly stock take finally turns up the gap.

1. Pull yesterday's closing figures for every product or category.
2. Compare them line by line against this morning's opening count — not just the total, the individual items.
3. For anything that doesn't match, check the obvious explanations first: an overnight delivery that wasn't logged, a return processed after closing, a miscount on either end.
4. If none of the obvious explanations account for the gap, flag it immediately rather than folding it into today's numbers and hoping it evens out later.

This is the exact gap Shelfie is built to close. An attendant photographs the shelf at the start and end of their shift on their own phone — no barcode scanner needed, and it works fine on 3G — and the app counts what's actually visible against what's expected, tied to whoever was clocked in for that shift (verified by face, so a shift can't be logged by someone who wasn't physically there). Instead of reconciling counts by hand every morning, the mismatch shows up flagged the same day, with a photo attached, before it has a chance to disappear into "we'll sort it out at month end."

What a real discrepancy investigation looks like

When the numbers genuinely don't add up, resist the urge to either ignore it or assume the worst. A proper investigation is boring and methodical, and that's exactly why it works:

1. Isolate the specific product or category that's short — a vague "something's missing" is not something you can act on.
2. Check it against actual sales for that shift. Did the volume sold explain most of the gap, or is there a genuine leftover discrepancy?
3. Check whether a delivery, return, or damaged-stock write-off happened during that shift and simply wasn't logged.
4. Identify who was on shift when the gap appeared — this only works if your counts are tied to a specific, verified person and not just "morning" or "evening."
5. Look for repetition. One mismatched count is very often a human error. The same product, the same size of gap, tied to the same shift or person across a week or more, is a pattern worth acting on.

The businesses that catch shrinkage early aren't the ones with the most suspicious owners — they're the ones with the most disciplined counting habits. A count taken at the right time, recorded properly, and reconciled the next morning turns a vague sense that "stock keeps disappearing" into a specific, provable answer.