Stocktake Planning

Plan a stocktake so the count is accurate, fast, and does not paralyse the business.

4 lessons 35 min 5-question assessment 75% to pass

What you’ll learn

Choose between full and cycle counts

Prepare properly for a count

Count accurately and avoid double-counting

Reconcile and explain variances

Lesson 1 of 4 Reading 8 min

Full vs cycle counts

A full stocktake counts everything at once, usually with the business closed. A cycle count counts a slice of stock regularly while trading continues. Both verify records against reality; they differ in disruption.

Choosing matters because a full count is thorough but stops the business and is rare, while cycle counts catch errors continuously without closing. Most operations blend a yearly full count with frequent cycle counts.

Closing your store for a full count of 5,000 SKUs might cost a day’s KES 800,000 in sales — worth it once a year for the books. But cycle-counting 200 high-value SKUs each week catches the KES 24,000 monthly shrinkage early without ever shutting the doors. The right mix depends on what disruption you can afford.

Key takeaways

Full count: everything at once, usually closed.
Cycle count: a slice regularly, while trading.
Most blend a yearly full count with frequent cycle counts.
Example: a full count may cost a day’s KES 800,000 in sales.

Lesson 2 of 4 Practice 9 min

Preparing for a count

Preparation makes or breaks a count: freeze stock movement during counting, ensure everything is in its location and labelled, assign counters to zones, and snapshot the system figure to compare against.

Preparing matters because counting while stock still moves guarantees errors — an item counted in aisle A then sold and recounted as missing wrecks the numbers. A clean freeze and clear zones make the count trustworthy.

Before counting, you stop receiving and dispatch, snapshot "system says 470 of this SKU", and assign counter 1 to aisles A–C and counter 2 to D–F so no shelf is counted twice or skipped. Without the freeze, a unit picked mid-count from aisle B appears as a phantom variance that you then waste an hour chasing.

Key takeaways

Freeze stock movement during the count.
Ensure stock is located, labelled, and zones assigned.
Snapshot the system figure to compare against.
Example: counting without a freeze creates phantom variances.

Lesson 3 of 4 Reading 9 min

Counting accurately

Accurate counting means counting each location once, scanning rather than eyeballing where possible, and recording the count against the location — not a running total in someone’s head. Blind counts (without seeing the expected figure) reduce bias.

Accuracy matters because the count becomes your new truth; a sloppy count replaces good data with bad and triggers wrong reorders and write-offs. The point of counting is to remove doubt, not add it.

A counter who can see "system expects 470" may subconsciously stop at 470; a blind count forces an honest 468, revealing a real 2-unit gap. Scanning each bin once and logging "B-04-A1: 60 units" prevents the classic double-count where a returning counter recounts a shelf already done.

Key takeaways

Count each location once; scan rather than eyeball.
Record counts against the location, not a mental total.
Blind counts reduce bias toward the expected figure.
Example: a blind count finds 468, exposing a real 2-unit gap.

Lesson 4 of 4 Reading 9 min

Reconciling variances

After counting, you reconcile: compare counted to system figures, investigate variances, recount big gaps, and only then adjust the records with a reason. A variance is a question to answer, not a number to blindly accept.

Reconciling matters because adjusting without investigating hides the real problem — booking a 30-unit loss as "shrinkage" when it was actually a mis-keyed receipt fixes the number but not the broken process behind it.

If the count shows 470 against a system 500, you recount to confirm, then check recent receipts and sales for that SKU before adjusting. You might find 20 units were a receiving mis-key (fix the process) and 10 are genuine shrinkage (a security flag). Each adjustment carries a reason, so next quarter’s variance tells a story instead of a mystery.

Key takeaways

Compare counted to system, investigate, recount big gaps.
Adjust records only with a recorded reason.
A variance is a question, not a number to accept blindly.
Example: a 30-unit gap may split into a mis-key and real shrinkage.

| AWRA OpsHub

Stocktake Planning

What you’ll learn

Course content

Full vs cycle counts

Preparing for a count

Counting accurately

Reconciling variances

Finished the material?

More in this path

Inventory Foundations

Warehouse & Locations

Units of Measure & Conversions

Need a quick answer while you read?

AwraIQ