Guides

How to run a cycle count in a warehouse

A cycle count checks a small part of your stock against the system, then moves on to another part next week. Over a few months you have counted everything that matters, and the warehouse never shut down for it.

The counting is the easy bit. The work is in choosing what to count and acting on what you find.

1. Know when a cycle count is the right tool

Use cycle counts to keep accurate numbers accurate. If the books are already close to the shelves, a few bins a week will catch drift before it grows.

If nobody has counted properly in years, start with a full stock take to get a clean baseline. Cycle counts on top of bad data mostly produce a long list of variances you cannot explain.

2. Choose what to count

You will not count everything every week, so pick the bins where a wrong number costs the most:

  • High-value items. A miscount of ten units matters more on a $400 part than a $0.20 washer.
  • Fast movers. More picks and receipts means more chances to get one wrong.
  • Bins that have had variances before. Errors cluster.
  • Bins that just had a large receipt, a return or a stock move.
  • Whatever a picker flagged as short.

3. Set a rotation

Write the schedule down so it happens without anyone deciding each week. A common pattern: count your top-value bins every week, your mid-range bins each month, and everything else once or twice a year. Another is one zone per day, working round the site.

Start with a pace your team can sustain, and tighten it where variances keep turning up.

4. Count without stopping the site

You are counting a handful of bins, so you only need those bins to hold still. Count early in the shift, before the day's picks start, or in a quiet gap. If someone needs stock from a bin you are counting, finish that bin first.

A count taken mid-pick will disagree with the system for reasons that have nothing to do with accuracy.

5. Count blind, then recount the differences

Counters should record what is on the shelf without seeing the system quantity. People round toward a number they can see.

When a count disagrees with the system, have a different person count that bin again before you conclude anything. Many variances are just a slip on the first pass.

6. Find the cause, then adjust

Check the recent history of the bin before you change its quantity:

  • Was something received against the wrong item?
  • Was a pick booked to a different SKU?
  • Did stock go into a neighbouring bin?
  • Was it counted in cartons and stored in singles, or the other way round?

Fix the process that caused it, then post the adjustment with a reason. If the numbers were badly off across the site, the stock take guide covers the usual causes in more detail.

7. Track accuracy over time

For each count, note how many bins matched the system on the first count. That percentage is the number to watch. If it climbs, your process is tightening. If one zone keeps dragging it down, look at how that zone receives and picks.

Where Varify fits

Start a session from New Counting Session and choose Cycle Count. There is nothing to configure: count the bins you picked, and only the items you count are checked against expected stock. Anything you skip stays out of the results. A stock take works the other way, where every item not counted shows up as a variance.

Counters scan the QR label on a bin with a phone. A second count of the same item becomes a recount instead of overwriting the first, and a Team Leader or above commits the result. A site can have one open session at a time, so finish one before you start the next. The full walkthrough is in How It Works. To load your bins and items first, see importing from Excel or CSV.

Try a cycle count on your own stock

Import your items and bin locations, then count a zone from your phone.

Start 7-day free trial

Monthly plans only. Card required. Trial sessions take up to 10 items.