A stock count is only as good as the sheet it is written on — and the discipline around it. This template carries the eleven columns a defensible count needs: where the stock was, what it was, what the system expected, what was actually there, who counted it and who checked. Used properly, the signed sheets are also a legal record: companies dealing in goods must keep statements of stocktakings behind their year-end stock figures under Companies Act 2006 s.386(4).
What each column is for
| Column | Why it is there |
|---|---|
| Location | The site, store or warehouse area being counted — one sheet per location keeps responsibility clean |
| Bin | The shelf, bay or bin reference. Counting by bin (not by SKU) is what catches stock sitting in the wrong place |
| SKU | The stock-keeping unit code exactly as it appears in your system — the join key for the reconciliation |
| Description | Human-readable item description, so counters are not matching codes blind |
| UoM | Unit of measure (EA, BOX, KG…). Most “losses” at reconciliation are really cartons counted as eaches |
| System Qty | Book quantity at the freeze timestamp. Hide this column for blind counts |
| Counted Qty | What was physically counted — written at the bin, never from memory |
| Variance | Counted minus system, calculated after sign-off. The investigation queue, in number form |
| Counted By | Who counted the line — accountability per line, not per sheet |
| Checked By | Who independently recounted or verified — never the same person as Counted By |
| Date | Count date (DD/MM/YYYY) — essential when counts span several days and movements continue |
How to run the count
- Freeze the book position. Extract system quantities at a recorded timestamp. Movements after the freeze are logged separately and reconciled as cut-off items — mixing them into the count is the commonest way counts go wrong.
- Prepare sheets by location and bin. One sheet per area, walked in physical sequence. For blind counts, print without the System Qty column.
- Count at the bin, write at the bin. Quantities recorded where they are counted, in the stated unit of measure. Damaged, expired or obsolete stock is counted and flagged, not skipped.
- Check independently. A second person recounts sampled lines and every line that looks unusual, signing the Checked By column. Variance thresholds trigger mandatory recounts.
- Reconcile before adjusting. Calculate variances, investigate causes — location errors, unit-of-measure mix-ups, cut-off timing — and only post what survives investigation, with authorisation appropriate to the value. That discipline is stock reconciliation, and it is where count data becomes a trustworthy number.
- File everything. Signed sheets, the frozen extract, the movement log and the adjustment schedule together form your statement of stocktaking — keep them with the accounting records.
When the spreadsheet stops being enough
This template will run an honest count for a stockroom, a small warehouse or a single store. The practical ceiling arrives with scale: thousands of SKUs, multiple sites counted to one standard, auditor attendance under ISA (UK) 501, or recurring variances that need investigating rather than recording. That is the point at which scan-based counting, controlled cut-offs and independent teams pay for themselves — the service described under stocktaking services, with cycle counting programmes for continuous accuracy and independent stock audits where the count must stand as third-party evidence. If your problem is equipment rather than stock, the equivalent free download is our fixed asset register template.