What a stocktake is actually for
A stocktake is not an inventory list. It is a comparison. The useful output is not "we have 43 cartons of milk"; it is "we have 43 and the records said 51, so eight have gone and we need to know how". Counting without something to count against produces a number and no information.
That comparison is where the money is. The gap between what should be on the shelf and what is on the shelf is theft, spoilage, breakage, mis-keyed sales, deliveries short-counted at the gate, and items sold at the wrong price. Each has a different fix, and you cannot pick the fix without first seeing the gap.
This is also why the annual count is weak. By the time you find a gap twelve months wide, the trail is cold: you cannot tell which month it happened in, which member of staff was on, or which supplier short-delivered. A gap found in one section this week is traceable.
Want to see it running in your own business first?
Sign UpCounting in sections
Divide the shop into roughly as many sections as there are weeks in a quarter. Each week, one section gets counted. The whole shop is covered every quarter and no day is ever a counting day.
- 1
Step 1: Draw the sections on paper first
Walk the shop and write down the sections: the drinks fridge, the dry goods shelf, the cleaning aisle, behind the counter, the store room. Number them. Fifteen to twenty sections is normal for a duka and gives you a full cycle each quarter.
- 2
Step 2: Count at a quiet hour, not after closing
Mid-morning on the slowest weekday beats 9pm after a full shift. Counting tired is where most errors come from, and a count with errors in it is worse than no count because you will chase differences that are not real.
- 3
Step 3: Count what is there, then look at the record
This order matters. If you read the expected number first you will see it on the shelf, which is a well-documented habit of the human eye and not a character flaw. Count blind, write it down, then compare.
- 4
Step 4: Investigate only the differences worth the time
Rank the gaps by shilling value, not by count. Two missing cartons of cooking oil matter; eleven missing sweets do not. Chase the top three and let the rest aggregate into a shrinkage figure you watch over time.
- 5
Step 5: Correct the record to the count, and note why
After investigating, make the system match reality and write the reason next to the adjustment. The reasons, accumulated over a quarter, are the actual report. "Adjusted: 14 units" tells you nothing; "adjusted: 14 units, expired" tells you to order less.
- 6
Step 6: Keep the rota even when nothing is wrong
The temptation after three clean sections is to stop. The rota is cheap insurance and its deterrent value, once staff know any shelf might be counted this week, is larger than anything it finds.
Why stocktakes go wrong
Counting everything at once
A full count is a day of work that produces one data point a year. The same hours spread across a quarter produce a continuous picture and catch problems while they are still small.
Letting the person responsible for a section count it
This is not an accusation; it is that nobody counts their own work neutrally. Rotate who counts which section, and the figures get better without anybody being accused of anything.
Counting with the expected figure visible
Showing the counter what the number should be turns a count into a confirmation. Print the sheet without the expected column, or count on paper and compare afterwards.
Adjusting the record without recording why
An adjustment with no reason destroys the only information the count produced. The quantity is the easy part; the reason is the part worth the exercise.
Counting during a delivery
Stock arriving mid-count is the most common source of a phantom gap. Count before the delivery or after it is fully received and booked in, never across it.
A worked example: a general shop in Nakuru
A shop with roughly 800 lines splits into 16 sections. Each Tuesday at 10am, one section is counted by whoever is not normally responsible for it. The count takes 25 to 40 minutes.
In week three, the cooking oil and fats section comes up eight units short across three sizes, about KES 4,400 at cost. The gap is recent, because the section was clean eleven weeks earlier, which narrows the question usefully.
The investigation takes twenty minutes. Two units were sold at the smaller size price, which the till receipt showed. Four were a short delivery the shop had signed for without counting. Two are unexplained and go to shrinkage.
Three different fixes come out of one count: a shelf label corrected, a rule that deliveries are counted at the gate before signing, and a KES 1,100 unexplained figure that is small enough to live with and now tracked. An annual count would have produced a single number twelve months later with none of this attached to it.
Stock you cannot see is stock you lose: dead capital sitting on slow shelves, empty shelves on your fast movers, and shrinkage no one can explain.
Veira tracks every item in and out with reorder alerts, so you hold the right stock and losses surface early.
Where the expected number comes from
Section counting only works if something is keeping a running expected quantity for you. That is the half of the job a shop cannot do on paper: every sale, every delivery and every adjustment has to move a number, continuously, without anyone remembering to do it.
Veira keeps that running figure as sales and deliveries are recorded, so a count sheet is a list of what should be there and a blank column. The adjustment and its reason are recorded against the item, which is what turns a quarter of counts into the reasons report described in step 5.
The counting itself is still a person with a shelf. No system can tell you what is physically on it, and any that claims to is guessing.
Frequently asked questions
How often should a small shop count stock?
Do I have to close the shop to do a stocktake?
What is an acceptable level of stock difference?
Should I count by value or by quantity?
Who should do the counting?
What do I do with the differences I find?
Can I do a stocktake without a system?
A stocktake earns its keep through the gap it reveals, not the list it produces, and a gap is only traceable while it is fresh. Count one section a week, count blind, and write down why you adjusted. See how Veira keeps the expected quantity current so the count has something to be compared against.
Need help, or want to talk it through first? Chat with us on WhatsApp