---
title: "Counting stock without closing the shop for a day"
source: https://www.lskhata.com/blog/counting-stock-without-closing-the-shop
category: "Stock & money"
language: en
published: 2026-08-28
reading_time: 6 min
---

# Counting stock without closing the shop for a day

_Full stock-takes get postponed until they never happen. Counting a few items at a time, continuously, catches more and costs nothing._

## In short
- Count ten items a day rather than everything once a year. The annual count is the one that never happens.
- Count the valuable and fast-moving items often; count the rest rarely or never.
- Every adjustment needs a reason recorded, or you learn nothing from having counted.
- Count before you look at the expected number. Reading it first changes what you count.

The traditional answer to stock accuracy is an annual stock-take: close for a day, count everything, reconcile. It works, and almost nobody does it, because closing for a day is expensive and there is always a better week next month.

The alternative is to stop treating counting as an event. Count a handful of items every day, during quiet minutes that already exist. Over two months you will have covered everything that matters, without ever closing, and you will have found each problem within days of it happening rather than eleven months later.

_[Figure: Ten items a day, on the days you are open. Nothing closes, and no single day is a burden.]_

### What to count, and how often

| Kind of item | How often | Why |
| --- | --- | --- |
| Expensive, fast-moving | Weekly | The most money at risk, and errors build up fastest. |
| Expensive, slow-moving | Monthly | High value, but fewer transactions to introduce mistakes. |
| Cheap, fast-moving | Monthly | Errors accumulate, but each one costs little. |
| Cheap, slow-moving | Twice a year, or not at all | The effort of counting genuinely exceeds what a difference would cost you. |

**Do not count everything equally.** Treating a two-hundred-rupee item with the same rigour as a two-rupee one is how counting becomes a chore that gets abandoned. Concentrating on the top of that table gives you most of the benefit for a fraction of the effort.

### The ten-minute routine

1. **Pick the items before you open, or the day before** — Ten items, chosen from your rotation rather than by wandering the shop and picking what catches your eye. Deliberate selection is what makes the coverage even.
2. **Count what is physically there, everywhere it is** — The shelf, the back, the box behind the counter. Stock in three places counted in one is the most common cause of a false difference.
3. **Write the count down before opening the app** — This is the step people skip and it is the one that keeps the count honest. Comparing after committing to a number is a genuine check; comparing while counting is not.
4. **Adjust, with a reason** — Damage, theft, sold under a different name, delivery never received. Record which. The reason is what turns a correction into information.

### Reading the reasons over time

A single difference is noise. A pattern in the reasons is worth acting on, and it only becomes visible if the reasons were recorded rather than the quantities silently corrected.

- The same item short every month, in small amounts: usually a billing problem — sold under the wrong product, or a multi-pack rung up as a single.
- One item short by a large amount, once: an event. Worth asking about while people still remember the week.
- Consistently over: often a receiving error, where more arrived than was recorded, or a return that was never entered.
- Differences clustered on one shelf or one shift: worth looking at as a pattern rather than as individual items.

**The movement history is the other half of this.** When a count does not match, the inventory movement history shows what the app thinks happened to that item and when. That is usually enough to find the mis-keyed bill or the unrecorded receipt without any detective work at all.

> The annual stock-take finds what went wrong. Counting a little at a time finds it while somebody still remembers why.

## Common questions

**How often should a small shop count stock?**

Continuously, in small amounts, rather than all at once. Ten items a day covers a few hundred products over a couple of months without ever closing the shop, and it finds problems within days of them happening rather than at the end of a year.

**Which products should I count most often?**

The expensive ones and the fast-moving ones. A difference on a high-value item costs the most, and a fast-moving item accumulates errors quickest. Slow, cheap items can be counted once a year or genuinely never.

**What do I do when the count does not match?**

Record the adjustment with a reason. The reason is the valuable part — damage, theft, a mis-keyed bill, an item sold under the wrong name. A quantity corrected with no explanation fixes today's number and teaches you nothing, so the same difference reappears next month.

**Should I count before or after looking at the expected figure?**

Before, always. If you know the system says 24, you will find 24 — not through dishonesty, but because counting is a task the brain is happy to finish early once it has a target. Count first, then compare.