Running an Annual Physical Inventory Count Without Chaos
An annual physical count exists to answer one question: does what's actually on the shelf match what your books say is there? The gap between those two numbers is shrinkage, and how you explain it, not just how big it is, is what an auditor or lender actually cares about.
A count that gets rushed on a Saturday with whoever's available almost always produces a number nobody trusts, including you. A count that's planned like a project produces a variance you can actually stand behind.
Cycle Counts vs. the Full Annual Count
A full physical count freezes operations and counts every SKU on one day, which is disruptive but gives you one clean number to reconcile against the general ledger. Cycle counting spreads counts across the year, a subset of SKUs at a time, so operations never fully stop, but it only works if you're disciplined about counting every location on a real rotation, not just the easy ones. Most small and mid-sized companies run cycle counts through the year for early warning and still do one full annual count near year-end to produce the number their auditor or lender actually relies on.
Who Counts and Who Reconciles Cannot Be the Same Person
The person who counts inventory should not be the same person who adjusts the books to match the count. If your warehouse lead both counts the shelf and posts the adjusting entry, an error, or a deliberate one, never gets a second set of eyes. Have warehouse staff do the physical count in pairs (one counts, one records), and have someone in finance who wasn't on the floor that day post the resulting adjustment after reviewing the variance report line by line.
Building the Count-Day Runbook
Freeze receiving and shipping for the count window, or clearly cut off and segregate any inventory in transit so it isn't double-counted or missed entirely. Print count sheets by location rather than relying only on a scanner app, since a second physical method catches scanner errors. Recount any SKU where the physical count differs from the system count by more than a small tolerance you set in advance, before you accept the variance as real.
- Freeze or clearly cut off receiving and shipping during the count window
- Count in pairs: one counts, one records, independently where possible
- Recount any material variance before accepting it
- Have finance, not the counting team, post the final adjusting entry
Explaining the Variance, Not Just Reporting It
A shrinkage number with no explanation reads as a control weakness even if the dollar amount is small. Break the variance into categories before you report it: known damage or spoilage written off during the year, receiving errors where the system quantity never matched what actually arrived, and unexplained loss, which is the category that should worry you most because it can mean theft. Gross margin by industry gives you a useful gut check here1: if your unexplained shrinkage is eating meaningfully into a margin that should otherwise track your industry's typical range, that's worth escalating before your next count, not waiting to explain again next year.
What to Fix Before Next Year's Count
If this year's count turned up a large unexplained variance, don't just book the adjustment and move on. Check whether receiving is verifying quantities against the purchase order before putting stock away, whether damaged goods are being written off as they happen instead of piling up and getting discovered only at count time, and whether your perpetual inventory system access lets warehouse staff make quantity adjustments without a second approval. Fixing the process that created the variance is worth more than getting a cleaner number next year by accident.
What an Auditor Actually Wants to See Afterward
Keep the raw count sheets, not just the final reconciled number, along with a short memo explaining every material variance by category and who approved the adjusting entry. An auditor testing inventory will usually pick a sample of locations or SKUs and ask to trace them back to the original count sheet, so if that paper trail doesn't exist, the count itself becomes a finding regardless of how accurate the final number turned out to be. File this documentation with the same discipline you'd apply to any other year-end closing binder, since inventory is often one of the largest balance sheet accounts an auditor will test.
What Good Looks Like
A good physical count process separates counting from recording, explains shrinkage by category rather than reporting one lump variance, and feeds fixes back into receiving and floor controls before the next count.
Building The Capability (5-Stage Skill Ladder)
How to Get Started
Frequently Asked Questions
How much inventory variance is normal versus a red flag?
There's no universal number: what's normal depends heavily on your product type, how perishable or theft-prone it is, and your industry's typical margin structure. The more useful test is whether you can explain the variance by category (damage, receiving error, unexplained loss) rather than comparing to a generic benchmark.
Do we need a full annual count if we already do cycle counts?
Most auditors and lenders still want one clean, full count near year-end to reconcile against the balance sheet, even if you cycle count through the year. Cycle counts catch problems early; the annual count is the number that ends up in your financial statements.
Who should physically perform the count if we're a small team?
Use whoever knows the warehouse, but pair them so the counter and the recorder are different people, and make sure the person posting the final adjustment in the books wasn't on the count team. If you're truly a one-person warehouse, have the owner or a fractional controller review and re-verify a sample of counted locations.
Sources
Where we quote a benchmark, we show its source. Other figures in this guide are estimates or general guidance, so check them against your own numbers.
- Gross margin by industry (US). NYU Stern (Aswath Damodaran), Operating and Net Margins by Industry, US, 2026.
Related Guides
Building a CFO Calendar That Never Misses a Deadline
How to build a single calendar covering every recurring tax, audit and compliance deadline, so nothing gets missed because it wasn't anyone's clear job.
FloQast vs. AuditBoard for Multi-Clinic Physical Therapy Networks
Contracted payer rates, visit-based billing, and claim denials complicate a physical therapy network's close. See how FloQast and AuditBoard fit.
Surviving a State Department of Revenue Audit
The audit methodology is often more negotiable than the assessment it produces. Here's the state tax audit process end to end, and when to push back.
Sales Tax Audits in California, New York and Texas
What makes a sales tax audit in California, New York or Texas different, and the records you need ready before an auditor's letter arrives.
How Long to Keep Corporate Tax and Audit Records
A practical records retention schedule for tax and audit documents, why the real minimum is longer than people assume, and how to actually enforce it.
How to Pass a Worker Classification Audit
The tests regulators actually use to decide contractor versus employee status, the red flags that trigger a look, and how to prepare before they come knocking.