Stock control for supermarkets: what to fix first

A supermarket is a stock control problem wearing a shop for a disguise. You might carry four thousand lines, and on most of them you are fine — the money leaks out of a few dozen you could name if you had the numbers in front of you. The trouble is that most stock management advice is written for a warehouse with a receiving bay and a systems team, and a Zimbabwean supermarket has a storeroom, a busy Saturday, and whoever is free to count.

So here is the order we would fix things in, based on what we see when we walk into a shop that knows something is wrong but not what.

1. Know what you actually hold

Not what the system says. Nothing downstream works until on-hand is real: reorder levels fire at the wrong time, valuation is fiction, and every variance report argues with itself. A full count is the reset, and after that the number stays honest only if every movement — sales, breakage, transfers, adjustments — goes through the same book.

2. Put reorder levels on the fast movers only

Setting a reorder point on four thousand lines is a project nobody finishes. Setting one on the two hundred lines that make your turnover is an afternoon, and it catches the stockout that actually costs you a customer. Bread, mealie-meal, cooking oil, sugar, airtime — you know the list; the system just needs to know it too.

3. Count the money, not the units

Two hundred missing units of something that costs twelve cents is a rounding error. Six missing units of something that costs forty dollars is a problem. Sorting variance by value rather than by quantity changes which shelf you walk to first, and it is the single fastest way to make a count worth the hours it took.

4. Separate breakage from shrinkage

Breakage is a cost of doing business — bottles get dropped, packaging gets crushed, dairy goes off. Shrinkage is stock you cannot account for. If breakage is not being recorded as it happens, it shows up in your variance as mystery loss, and you end up investigating theft that is really a wet floor by the fridges.

5. Count more often, and smaller

The annual count exists for the auditor. It is close to useless for running a shop, because it tells you in March about something that happened in September. Cycle counting — a section a week, high-value lines more often — finds problems while they are small enough to trace to a delivery or a shift.

What this looks like when it works

• On-hand you would bet money on, because every movement goes through one book
• Reorder alerts on the lines that matter, and silence on the ones that do not
• Variance sorted by value, so the first shelf you check is the expensive one
• Breakage recorded where it happens, so shrinkage means what it says
• A count that takes a morning instead of closing the shop for a weekend

None of that needs a warehouse management system. It needs one place where stock lives, a till that updates it as it sells, and a way to count that does not involve a clipboard and a calculator.

“Stock control is not about counting everything. It is about knowing which things are worth counting often.”