Audit Sampling: Definition, Methods & Examples

What Is Audit Sampling?

Audit sampling, as defined by International Standard on Auditing 530, applies audit procedures to fewer than all items within a population of audit relevance, provided every sampling unit has a chance of being selected. The purpose is to allow the auditor to evaluate evidence about a characteristic of the population and form a conclusion about the whole.

Sampling connects materiality with audit evidence. Materiality sets the threshold for what would matter to users of the financial statements, while sampling determines which items the auditor examines to test whether that threshold may have been breached. The design of the sample therefore shapes where audit effort is concentrated and how persuasive the resulting evidence will be.

How Audit Sampling Works

The process begins with the audit objective. An auditor may want to confirm whether recorded receivables exist, whether payables are complete, or whether a control operated throughout the period. That objective determines the population from which the sample is drawn, because a sample can only support the conclusion if the population matches the assertion being tested.

Sample size is then driven by the auditor's assessment of risk. A lower tolerable misstatement, a higher expected error rate, or a higher desired level of confidence will usually require more testing. Population structure also matters, since a population containing a small number of high-value items may need to be divided into bands so that individually significant balances receive direct attention.

Selection methods vary according to the engagement. Random selection gives every item an equal probability of selection, systematic selection picks every nth item after a random start, stratified sampling divides the population by value or characteristic, and monetary unit sampling gives larger recorded balances a greater chance of selection. After testing, the auditor projects any misstatements found across the relevant population and assesses whether the projected error, together with known errors, exceeds the tolerable threshold.

Audit Sampling Example

Consider an external auditor testing year-end trade receivables for a UK distributor with 4,200 outstanding invoices totalling £18 million. The auditor separates the population into three bands. All 35 invoices above £100,000 are tested individually, 60 invoices between £20,000 and £100,000 are selected using monetary unit sampling, and 40 invoices below £20,000 are chosen at random.

Testing identifies two overstatements in the middle band totalling £14,000, which the auditor projects across that band to estimate a total population misstatement of £58,000. Because the projected amount remains below the tolerable misstatement threshold of £90,000, the result supports the auditor's conclusion on receivables. The engagement tested fewer than 4 percent of total invoices, yet the reliability of the conclusion depends on whether the sampling design captured the population's actual risk profile.

Key Considerations and Limitations

Audit sampling produces reliable evidence when it is proportionate to the risk profile of the population being tested, although it always carries sampling risk. Incorrect acceptance is the more consequential form for users of financial statements because the sample may suggest that no material misstatement exists when the population is misstated. Increasing sample size reduces that exposure, but it cannot remove it completely.

Non-sampling risk sits outside the mathematics of selection. An auditor may choose an appropriate sample but apply the wrong procedure, misinterpret evidence, or fail to investigate an exception properly. Testing more items does little to solve that problem, which is why audit quality depends on both sample design and the professional discipline applied during testing.

Audit committees overseeing external audit quality under the UK Corporate Governance Code should be able to challenge whether the auditor's sampling approach was aligned with the financial reporting risks in the engagement. The governance question is practical rather than technical, because directors need assurance that the audit work was targeted at the areas where misstatement would matter most.

Audit Sampling vs Analytical Procedures

The limits of sampling raise a practical question about when auditors should use another evidence-gathering technique. Analytical procedures, governed by ISA 520, evaluate financial information by studying relationships, trends, and ratios rather than testing selected individual items. The distinction affects how auditors allocate effort across the engagement.

When the risk of material misstatement is high and the auditor needs transaction-level evidence, audit sampling is usually the stronger technique. When the risk is lower and the objective is to confirm that a balance is broadly reasonable, analytical procedures may provide a more efficient route to the same assurance objective.

In Practice

Audit sampling is a bridge between detailed transaction testing and the broader assurance that directors, investors, and lenders rely on when reading audited financial statements. Its value depends on the clarity of the test objective, the completeness of the population, the appropriateness of the selection method, and the quality of judgement applied when exceptions are evaluated.

For executives and audit committees, the practical issue is whether the auditor's work has addressed the areas where error or fraud would have the greatest reporting consequence. A technically valid sample can still be weak if it misses the commercial reality of the account being tested. Strong audit oversight therefore requires enough understanding of sampling to ask focused questions about risk, coverage, exceptions, and the basis for the final conclusion.