The plain-language guide
How do you count stock fast at close?
A close count that takes 45 minutes gets skipped by the third night. One that takes 5 minutes gets done. Here is what each kind of item costs in seconds, which items to count every night, and how to read the gap.
The short answer
Count the 10 items that hold most of your weekly spend every night, in shelf order, and rotate the rest in groups of about 8. In the 42-item stall below that takes 4.4 minutes a night, against 9.0 minutes for a full count. Time any count as items × seconds per item ÷ counters, plus the walk. Judge each gap by its value in pesos and its share of the record.
What the count protects
DeHoratius and Raman examined nearly 370,000 inventory records from 37 stores of one retailer and found 65% of them inaccurate. They also found that 26.4% of the variation in record errors sat between product categories, and 2.7% between stores. The type of item predicted the error better than the shop did. The study covers one retail chain with no restaurants in it, so the 65% is a warning about stock records in general and says nothing about the rate in your kitchen.
Take the effect to a kitchen. The record says 12.0 kg of chicken thigh and the shelf holds 10.5 kg. The next order starts from a number that is 1.5 kg too high, a 12.5% error on that item. If the supplier needs 2 days of notice, the shortfall shows up on the day the walk-in is empty and the delivery is still 2 days out.
How long a full count takes
The example stall is a burger counter with 42 stock items, closing at 22:00. The seconds per item below are example timings for this stall. Time your own on one close with a phone timer before you plan around them.
| Item type | Items | Seconds each | Seconds in total |
|---|---|---|---|
| Sealed packs, counted by eye | 12 | 6 | 72 |
| Boxes and cases | 14 | 10 | 140 |
| Loose items, weighed | 9 | 25 | 225 |
| Opened bags, estimated | 7 | 15 | 105 |
| Total | 42 | 12.9 average | 542 |
542 seconds is 9.0 minutes for one counter. Two counters, one calling and one writing, bring the counting to 4.5 minutes. The walk between the walk-in, the dry store and the line stays at about 2 minutes either way, so the full count takes 11.0 minutes alone and 6.5 minutes in a pair. Weighing is the expensive step: the 9 weighed items are 21% of the items and 42% of the seconds (225 of 542).
Time your own close
11.0
minutes for the whole count
Which items earn a count every night
Warehouses solve this with cycle counting: a rolling audit of a few items at a time, with the count frequency tied to value or usage. The published schedules disagree with each other, which is useful to know before you copy one.
| Source | What it sets |
|---|---|
| Wikipedia, Cycle count | Higher-value items are counted more often. Selections are generated daily, though many firms generate them weekly. Every location at least 1 time a year. |
| RF-SMART | Class A items quarterly (4 a year), class B twice a year, class C once a year. |
Both set the busiest class at 4 counts a year or fewer, and both are written for warehouses where stock keeps for months. A burger counter holds chicken for days. The Wikipedia article also names the weak point of ranking by value alone: a low-value item that is ignored can stop the line when it runs out. In a kitchen that is the burger wrapper or the gas canister.
Here is the ranking for the example stall. Its weekly purchases total ₱39,000. The top 10 of 42 items (24% of the items) account for ₱31,200 of that, which is 80%. Counting those 10 at 15 seconds each takes 150 seconds. The other 32 items are split into 4 groups of 8, one group a night, at 14 seconds each, which is 112 seconds. The nightly count takes 262 seconds, or 4.4 minutes, plus the walk. Each of the 32 slower items is still counted every 4th night.
- Rank every stock item by what you spend on it in a week.
- Add every item that stops a dish when it runs out, whatever it costs.
- Count that list every night.
- Split the remaining items into groups of about 8 and count 1 group a night.
- Count the whole shelf once at the start of each month, to catch items the groups missed.
The order to walk the shelves
Sort the count list by where each item sits. A list sorted by name sends the counter from the walk-in to the dry store and back, and every crossing adds to the 2 minutes of walking. A list sorted by location makes one pass through each room.
Count before you clean and before you restock the line. A bin of trimmed onions that is emptied first removes the evidence of the waste you are trying to find. Write each number as you count it. A figure held in memory for 6 shelves is the most common source of a transposed digit.
Reading the gap
The table shows 5 items from one close. The gap is the count minus the record. The value is the gap times the cost per unit.
| Item | Record | Counted | Gap | Cost per unit | Gap value | Share of record |
|---|---|---|---|---|---|---|
| Chicken thigh (kg) | 12.0 | 10.5 | -1.5 | ₱240 | -₱360 | -12.5% |
| Burger buns (pcs) | 96 | 90 | -6 | ₱6.50 | -₱39 | -6.3% |
| Cheese slices (pcs) | 150 | 150 | 0 | ₱8 | ₱0 | 0.0% |
| Cooking oil (L) | 20 | 21 | +1 | ₱110 | +₱110 | +5.0% |
| Beef patties (pcs) | 80 | 71 | -9 | ₱38 | -₱342 | -11.3% |
The net gap is -₱631. Set the rule before the count, so you know which gaps to chase before you see them. The example rule is to investigate any item whose gap is ₱300 or more, or 10% or more of its record. Chicken (₱360, 12.5%) and patties (₱342, 11.3%) meet it. Buns and oil do not. Set your own thresholds from your own margins.
The sign carries information. A minus on the same item every night points to portioning, unrecorded waste or an unrung sale. A plus points to a delivery that nobody received into stock. A gap that flips between plus and minus from night to night points to counting error, so count that item twice before looking for a cause.
What ForeFlux does
On the Inventory screen, the Adjustments tab has a Count action. Choose the item and its branch, enter the number you counted, and ForeFlux works out the change against the record. Pick a reason. Each correction is logged, so you can see why a number moved. Until you enter an opening count, Inventory reminds you that its figures are the stock ForeFlux expects, and that nothing has been counted yet. Every order suggestion starts from that stock figure, so a fresh count changes the next suggestion.
Sources
- DeHoratius, N. and Raman, A. (2008), Inventory record inaccuracy: an empirical analysis, Management Science 54(4). Nearly 370,000 records, 37 stores of one retailer, 65% inaccurate, 26.4% of variance between categories, 2.7% between stores.
- Wikipedia, Cycle count. ABC analysis with the Pareto method, value-based and usage-based counting, the low-value-item weakness.
- RF-SMART, How to develop a cycle count process. Class A items quarterly, class B twice a year, class C once a year. A vendor blog, written for warehouses.
- The stall, its 42 items, its timings and its pesos are an invented example for arithmetic. They come from no customer and no measured result.
Questions owners ask
Past the basics.
Do I count opened packs and sealed packs the same way?
Sealed packs are counted as whole units and cost about 6 seconds each in the example. Opened bags are estimated in the unit you order by, and a weighed estimate costs about 25 seconds. Weigh an opened item only when its value justifies it. A half-used 25 kg sack of rice at ₱50 per kg is worth ₱1,250, so weigh it. A half-used pack of cup lids is worth less than the 25 seconds.
What should I do when the count never matches the record?
Look at the sign of the gap over 5 closes. The same sign every night on one item has one cause to find. A sign that flips has no cause beyond counting error, and counting the item twice removes most of it. If the gap is still there after 2 careful counts, correct the record to the counted figure and write the reason down, so the next order starts from what is on the shelf.
Can I count at opening instead of at close?
Yes. A close count lines up with the day of sales you rang up. An opening count measures what happened overnight as well, including spoilage in the walk-in. Either one works if you do it at the same time every day, because the gap you read is always the change since the last count.
See the next order before you need it
Know what you need, before you need it.
The demo needs no signup and no card. It is the real product running on demo data.
Join the waitlist Open the order list