How to do a warehouse stock count
A warehouse stock count means counting the physical quantity of every product on the shelves, one by one in shelf order, comparing it with the quantity on record (in your software or Excel), and recording any difference together with its cause. A count that gives reliable results has three conditions: no goods move in or out of the warehouse during the count, the person counting doesn't see the system quantity (a blind count), and any product with a discrepancy is counted a second time.
The six steps below apply in the same order to a small business with a single warehouse and to a warehouse with several racking systems.
1. Choose the count day and stop movement
A count is done at a moment when the warehouse is at a standstill. Choose a time after closing or a weekend; stop goods receipt and shipping until the count is finished. Set aside any goods waiting on the dock outside the counting area and label them "not to be counted" — the most common cause of count discrepancies is goods coming in or going out while the count is under way.
2. Prepare the warehouse for counting
Consolidate shelves where the same product is spread out, close open cartons, and make sure shelf labels are legible. Divide the warehouse into zones by aisle-rack-slot order and give each zone to one person; that way no shelf is counted twice or not at all.
3. Prepare a blind count sheet
The sheet should have the product code, name, shelf address and an empty "counted quantity" column; don't write the system quantity on the sheet. A person who sees the system quantity tends to confirm it instead of counting. This method is called a blind count, and it is the single step that improves count accuracy the most.
4. Count the shelves in order
Go zone by zone, in shelf order; don't wander around the warehouse looking for products. For barcoded products, scanning with the phone camera is both faster and more accurate than writing by hand. If two people are working, one counting and one recording, have the quantity repeated out loud.
5. Find the discrepancies and have the mismatches recounted
Enter the counted quantities into the system and compare them with the system quantity. Have every product with a discrepancy recounted by someone other than the first counter. Most discrepancies are not real losses but goods placed on the wrong shelf or the same product opened under two different codes.
6. Record the cause of the discrepancy and correct the stock
After the second count, post the remaining difference to the system, but write down its cause first: breakage/shrinkage, wrong shipment, an unrecorded sale, theft. A correction without a recorded cause doesn't prevent the same discrepancy from appearing again in the next count. If part of the difference is explained as “it was broken, it just wasn't recorded”, the problem is not in the count but in the shrinkage record; we covered recording shrinkage on the day it happens in the what is shrinkage and how to keep a shrinkage record guide. Finally, put who did the count, on which days and at what times, together with the shortage/surplus result, into a document for the counting team to sign; there is a free stock count report example that fills in the result summary automatically from the product list.
Full count or cycle count?
The two methods are not alternatives to each other; they complement each other:
| Topic | Full count | Cycle (partial) count |
|---|---|---|
| Scope | All products in the warehouse | A selected product group or zone |
| Frequency | At least once a year | Monthly or weekly |
| Warehouse status | Movement stops | The warehouse keeps operating |
| What it is for | Actual year-end inventory value | Catching discrepancies before they grow |
| Cost | High: operations stop | Low: spread across the daily flow |
When choosing which products to start with in a cycle count, put high-value, fast-moving products first; they are what carries your revenue.
Calculate your stock accuracy rate
You can track whether counting is working with a single number:
Stock accuracy rate = (number of product types with no discrepancy ÷ total number of product types counted) × 100
If you counted 500 product types and 465 matched the records exactly, your accuracy rate is 93%. Write this rate down at every count; the goal is not to hit a particular figure but for the rate to rise from one count to the next. If the rate is falling, the problem is not the counting method but the daily receipt and dispatch records — review the setup of your warehouse inventory tracking from the beginning.
What makes counting easier: preparation
A count usually takes long not because of the counting itself but because of searching for products. Addressing the shelves, giving every product a unique code and defining reorder levels shorten count day. You can find the reorder level with the reorder level calculator, and if you will run the count in Excel, you can download the free warehouse count template.
If you do the count with software, the job gets shorter: you scan the barcode, the system quantity and counted quantity appear side by side, and the difference and its value are calculated. The screens are described on the features page, and for working in the warehouse with a phone, see the mobile and web versions.
Frequently asked questions
How long does a warehouse stock count take?
What determines the time is the number of product types, not the quantity. One person counting with a barcode scanner gets through roughly 150-250 product types per hour; a warehouse of 1,000 types is finished in half a day with two people. If shelf addresses aren't labeled, this time can double.
What is a blind count?
It means not showing the system quantity on the count sheet. The person counting writes down the quantity they see on the shelf without comparing it to a number they don't know; this prevents brushing it off with “it probably matches anyway”.
Full count or cycle count?
They are used together. A full count covering the whole warehouse is done at least once a year; for fast-moving, high-value products, a monthly cycle (partial) count catches differences in between before they grow.
How big a count discrepancy is normal?
Rather than an acceptable threshold, track your own measurement: if your accuracy rate rises with every count, the process is working. If the same product shows a discrepancy repeatedly, the problem is not the count but that product's receipt and dispatch records.
Can I do the count in Excel?
Yes. You can export the product list with code, name, shelf and counted quantity columns and fill it in by hand; there is also a ready-made warehouse count Excel template that calculates the difference with a formula. The limitation of Excel is this: two people can't work in the same file while the count is under way.
What should be done after the count?
After closing the discrepancies with their causes, recalculate your reorder levels based on the current sales rate, and put the date of the next cycle count in the calendar that same day. Because a count also pins down the value of the stock you hold, it is the best moment to calculate inventory turnover.
Last updated: 2026-09-25
The counting screen in the app
System quantity, counted quantity and difference side by side.
Do your first count with your phone
Open a free account, import your Excel list and scan barcodes as you walk the shelves.