GenFrontier homeScope a project
Menu
Filed under
Deliver toInventory control leads and warehouse managers running counts

Inventory accuracy calculator with count tolerances

Enter system and counted quantities with a tolerance per line to get inventory accuracy as ASCM defines it, plus unit and dollar variances, as a CSV.

Plan an inventory project
resources / inventory-accuracy-calculator
A spirit level resting across the tops of a row of identical cartons, its red bubble window showing the bubble exactly centered

Calculator

Five fictional count lines are filled in. Edit them, add your own lines and download the result as CSV. Nothing leaves your browser.

SKU or locationSystem quantityCounted quantityUnit cost ($)Tolerance (%)VarianceWithin toleranceRemove
0Yes
-2No
-1No
-12Yes
+3No

Results

Lines counted
5
Inventory record accuracy (lines within tolerance)
40.0%

The ASCM dictionary defines inventory accuracy as the condition “when the on-hand quantity is within an allowed tolerance of the recorded balance.” It “usually is measured as the percent of items with inventory levels that fall within tolerance,” and target values “usually are 95 percent to 99 percent, depending on the value of the item.” This calculator applies that definition to a list of count lines.

How to use the calculator

  1. Enter one line per count. A line can be an item, a location or an item in a location. Use the same choice for every count you want to compare.
  2. System quantity is the recorded balance at the moment of the count. Stop transactions in the location or record a cutoff time, otherwise you are counting against a moving number.
  3. Counted quantity is what the counter found. Recount before you enter a variance on a high-value line.
  4. Unit cost at cost, so the variance can be valued.
  5. Tolerance as a percentage of the system quantity. Zero means the count must match exactly.
  6. Add lines for your own counts, then download the CSV for your records or a root-cause review.

The formula

A line is within tolerance when:

|counted − system| ≤ system × tolerance ÷ 100

Inventory accuracy = lines within tolerance ÷ lines counted × 100

The calculator also reports the total absolute unit variance, the sum of every miss regardless of direction, and the net value of variances, the sum of each variance times its unit cost.

Example (fictional). Five count lines from one morning’s cycle count:

SKU / location System Counted Variance Tolerance Allowed Within
EX-1001 / A-01-02 240 240 0 0% 0 Yes
EX-1002 / A-03-01 96 94 −2 2% 1.92 No
EX-1003 / B-14-02 40 39 −1 0% 0 No
EX-1004 / C-07-04 1,200 1,188 −12 2% 24 Yes
EX-1005 / C-09-01 15 18 +3 0% 0 No
  • Lines within tolerance: 2 of 5, so inventory accuracy is 40.0%
  • Total absolute unit variance: 0 + 2 + 1 + 12 + 3 = 18 units
  • Net value: (−2 × $18.75) + (−1 × $22.00) + (−12 × $0.40) + (3 × $64.00) = +$127.70

Reading the result

The two summary numbers tell different stories. Finance sees a net variance of $127.70 in the company’s favor. Operations sees that three of five lines were wrong. The positive dollar figure comes from one overage on an expensive item, EX-1005, offsetting shortages elsewhere. Offsetting errors are still errors: a picker sent to B-14-02 for 40 units will find 39.

The miss on EX-1002 shows how percentage tolerances behave on small quantities. Two percent of 96 is 1.92 units, so a two-unit shortage fails. If your rule is really “two units either way”, set a tolerance that produces it, or keep it tight on purpose and say so.

Tolerances should follow value. ASCM ties the target to the value of the item, and most programs set zero tolerance for expensive or controlled items and a small percentage for cheap, high-count items such as EX-1004 at $0.40 a unit.

Common mistakes

  • Counting against a moving balance. Picks and receipts during the count create false variances.
  • Recounting until it matches. A second count is for confirmation, not negotiation.
  • Reporting only dollar accuracy. Net dollars hide offsetting errors.
  • Skipping empty locations. Counting locations the system says are empty is how misplaced stock is found.
  • Adjusting without asking why. ASCM describes the key purpose of cycle counting as identifying items in error, “thus triggering research, identification, and elimination of the cause of the errors.” Every adjustment should carry a reason code.

When the simple model stops working

  • Lot and serial tracking. The quantity can be right and the lot wrong, which matters for recalls and expiry.
  • Location accuracy. The building can hold the right total while individual bins are wrong. Measure at the level where people pick.
  • Unit of measure errors, where a case is counted as an each or the reverse.
  • Biased samples. Accuracy measured only on A items, or only on locations that were easy to reach, overstates the whole.

Plan the counting itself with the cycle count planner. If accuracy stays low after regular counting, the cause is usually in the process: receiving without scanning, moves that are never recorded, or adjustments that post days late. Finding and fixing those causes is the work of inventory management software and barcode, RFID and labeling projects.

Scope your project.

Tell us what you need to track, which systems you run and when it has to work. We reply with questions or a time to talk.