Glossary
Cycle counting
Counting a small part of the warehouse every day instead of the whole warehouse once a year, so errors are found while the trail is still warm.
What it is
Cycle counting is a schedule. Each day, or each shift, a few locations are counted and compared to the system. Over a cycle, a week, a month, a quarter, every location has been counted at least once, the important ones several times. The warehouse never stops to count everything, and the record never gets more than a cycle old.
Why it matters
A discrepancy found the day it happened can usually be explained: the pick from the wrong bay, the receipt keyed as cases instead of pieces, the return that never got put away. The same discrepancy found eleven months later is a number with no story, and the only fix is to overwrite it and hope.
Cycle counting also changes what a count means. It stops being an event and becomes evidence, a running proof that the ledger and the shelf agree.
A worked example
Tuesday’s cycle includes bay B-04-02. The counter scans the location, then scans ITEM-01 and enters a count of forty-six. The system expects forty-eight. The count is recorded as a variance of minus two, not silently applied. A supervisor opens the location’s movement history: a pick of two on Monday was confirmed but the picker scanned a neighboring bay. The picker’s row is the explanation; the adjustment row carries the reason “picked from B-04-02, recorded at B-04-03”, and both bays are corrected with names attached.
How DockingWare does it
Counts are documents in the ledger, not edits to a balance. A variance is recorded before it is applied, and an adjustment is its own movement row with a reason and a user. Because every balance is derived from rows, the history a supervisor needs to explain a variance is already there: the picks, receipts and transfers that touched that location, in order, with names.
Questions
- Does cycle counting replace the annual physical count?
- In most operations, yes. If every location is counted on a cycle and the variances are explained and corrected, the annual shutdown count stops finding anything the cycle did not already find.
- What should be counted first?
- Whatever hurts most when wrong: fast movers, high-value items, locations with recent variances, and anything a system flagged as suspicious. Counting slow, cheap stock every week is effort without return.
- What happens when the count disagrees with the system?
- An adjustment is written as its own movement, with a reason and the counter's name. The old number is not overwritten; a row says what changed and who said so.