Inventory Sample Size Calculator
An inventory sample size calculator tells you how many items to count in a stock audit to reach a chosen confidence level and margin of error, without counting everything. It uses the standard proportion formula with a finite population correction, so the answer stays honest for small warehouses as well as huge ones.
Assumes simple random sampling. Results are rounded up: you cannot count a fraction of an item, and rounding up keeps the promised confidence.
Statistical guidance for planning inventory audits. Valid results require genuinely random sampling; hand-picked or convenience samples do not carry the stated confidence. This is not professional audit advice.
How the calculator works
Enter the total population N, pick a confidence level (90%, 95%, or 99%), and enter your margin of error and expected proportion. The tool first computes the unadjusted sample size n₀ = (Z² × p × (1 − p)) ÷ E², then applies the finite population correction n = n₀ ÷ (1 + (n₀ − 1) ÷ N), and rounds up to a whole item. You get the final count, the sampling percentage, and a plain-English reading of what the numbers promise.
The formula, explained
The Z-score encodes your confidence: 1.645 for 90%, 1.96 for 95%, 2.576 for 99%. The proportion p is your best guess at the error rate you are measuring; p = 0.5 gives the largest, most conservative sample because variability peaks at 50/50. The margin of error E must be a decimal in the formula (5% → 0.05). The finite population correction matters when the sample is a noticeable fraction of the population: without it, a 200-item warehouse at 99% confidence would demand an absurd 1,844 counts instead of the correct 181.
Why this beats the old tool
The previous version of this page computed the same formula correctly, but it buried the answer in popup alerts, auto-recalculated on every keystroke before inputs were complete, and showed a decorative “efficiency” gauge instead of the working. The rebuilt tool keeps the verified formula identical, shows every arithmetic step so an auditor can check it, validates inputs inline, and states the result as a sentence you can put in an audit plan.
Worked example, verified against the tool
A warehouse holds 5,000 SKUs. You want 95% confidence, 5% margin of error, and assume maximum variability (p = 0.5). Z = 1.96, so n₀ = (1.962 × 0.5 × 0.5) ÷ 0.052 = (3.8416 × 0.25) ÷ 0.0025 = 384.16. Finite population correction: 384.16 ÷ (1 + 383.16 ÷ 5,000) = 384.16 ÷ 1.07663 = 356.82, rounded up to 357 items, which is 7.14% of the population. Every figure matches the tool exactly.
When to sample and when to count everything
Sampling wins when the population is large and counting everything costs more than the risk it removes. As a rule of thumb, when the required sample exceeds about half the population, just count everything: the savings no longer justify the sampling risk. Stratify first when the inventory has distinct groups (high-value vs. low-value SKUs); sampling each stratum separately is more precise than one blended sample.
Frequently asked questions
How many items should I count in an inventory audit?
It depends on your population size, confidence level, margin of error, and expected error rate. For 5,000 items at 95% confidence and 5% margin of error, the answer is 357 items. Enter your own numbers above.
What is the sample size formula for inventory auditing?
n₀ = (Z² × p × (1 − p)) ÷ E², then n = n₀ ÷ (1 + (n₀ − 1) ÷ N) for the finite population correction, rounded up.
What should I use for the expected proportion p?
Use 0.5 unless you have a solid prior estimate. It maximizes the required sample, so your audit stays valid even if the true error rate differs from your guess. If last year’s audit found a 2% discrepancy rate, p = 0.02 gives a smaller, still valid sample.
Why round the sample size up?
You cannot count a fraction of an item, and rounding down would drop you below the promised confidence level. Rounding up is the conservative, correct choice.
Does the sample have to be random?
Yes. The confidence statement only holds for random samples. Counting the easiest-to-reach shelf, or the items you already suspect, invalidates the math.
What margin of error should I choose?
3–5% is typical for inventory audits. Tighter margins demand much larger samples: halving the margin of error roughly quadruples the sample size.
Why does a higher confidence level need a bigger sample?
More certainty requires more evidence. Moving from 95% to 99% confidence raises the Z-score from 1.96 to 2.576, which increases the unadjusted sample size by about 73%.
When should I skip sampling and count everything?
When the population is small (a few hundred items or fewer) or the required sample approaches half the population, full enumeration is usually cheaper than the risk and effort of a defensible sample.