toolbox
Home / Legal, Safety & Compliance / Vendor Compliance Audit Sample Size Estimator

Vendor Compliance Audit Sample Size Estimator

The Vendor Compliance Audit Sample Size Estimator calculates a planning sample size for a finite vendor population when the objective is to estimate a proportion, such as the share of vendors meeting a defined control or documentation criterion. It uses a standard proportion sample-size model with a finite-population correction, based on population size, confidence level, margin of error, and an expected compliance proportion.

This is useful during audit or monitoring design when teams need a transparent statistical starting point before selecting actual records. A 50% expected proportion produces the most conservative sample for a given confidence level and margin of error because it maximizes estimated variability. The result does not decide which vendors should be selected, whether sampling is appropriate for a specific audit objective, or whether a regulator, contract, assurance standard, or audit methodology requires a different approach. Those decisions should be set by the responsible audit or compliance function.

Inputs

vendors
%
%
Result
—
Estimated sample size
Uncorrected sample—
Sample as % of population—
Population—

1. Enter the vendor population
Use the number of vendors or vendor records in the population from which the sample will be drawn.

2. Select a confidence level
Choose 90%, 95%, or 99% according to the planning standard used for the review. Higher confidence generally requires a larger sample.

3. Set the margin of error
Enter the desired precision in percentage points. A smaller margin of error increases the required sample.

4. Estimate the expected proportion
Enter the proportion expected to meet the tested criterion. Use 50% when no reasonable prior estimate exists and a conservative sample is preferred.

5. Use the rounded-up sample
The result is rounded up to a whole vendor and capped at the population size. Apply the audit team’s approved selection method separately.

Initial sample n₀ = z² × p × (1 − p) ÷ e² Finite-population sample n = n₀ ÷ [1 + (n₀ − 1) ÷ N]

N is the vendor population, z is the z-score for the selected confidence level, p is the expected proportion expressed as a decimal, and e is the desired margin of error expressed as a decimal. The displayed sample is rounded up to the next whole vendor and cannot exceed N.

This model estimates a proportion under conventional random-sampling assumptions. Risk-based, stratified, discovery, monetary-unit, control-testing, or regulator-prescribed samples may require a different method.

What the result means

The result is a statistical planning estimate of how many vendors to sample for a proportion estimate under the assumptions entered.

Sampling design, selection method, tolerable deviation, expected deviation, and audit standards may require professional judgment beyond this estimator.

Given
2,400 vendors, 95% confidence, a 5% margin of error, and a 50% expected compliance proportion.

Calculation
n₀ = 1.959964² × 0.50 × 0.50 ÷ 0.05² = 384.15.
n = 384.15 ÷ [1 + (384.15 − 1) ÷ 2,400] = 331.35.

Result
Round up to 332 vendors.

Interpretation
Under the stated statistical assumptions, a simple random sample of 332 vendors is the planning sample for estimating a proportion with the selected confidence and precision.

Why does a 50% expected proportion give a larger sample?

For a proportion estimate, variability is highest at 50%. When no better estimate is available, 50% therefore provides a conservative sample size for the same confidence level and margin of error.

Should my population include inactive vendors?

Include only records that belong to the population defined by the audit objective. If inactive vendors are outside scope, exclude them before entering the population size.

Is the calculated sample automatically audit-compliant?

No. An audit methodology, regulator, contract, or assurance standard may prescribe a different sample design or require additional judgment. Treat this as a statistical planning estimate, not a substitute for the applicable audit approach.

Can I use the result for a risk-based sample?

Not directly. Risk-based sampling intentionally changes selection probabilities or focuses on higher-risk items, while this estimator assumes a proportion estimate under conventional random-sampling conditions.

Why does the sample stop growing as a share of a large population?

Once a population is large relative to the sample, additional population size has limited effect on the precision of a proportion estimate. The finite-population correction matters most when the sample is a substantial share of the population.