ISA 530 audit sampling: a summary of the standard
ISA 530 is the IAASB standard on audit sampling. It covers how to design the sample, set its size, select the items, deal with deviations and misstatements, project them and evaluate the result, so that the sample gives you a reasonable basis for conclusions about the whole population. It sets no formulas and no minimum sizes: the sample must reduce sampling risk to an acceptably low level.
Updated 10 November 2026
What is ISA 530?
ISA 530 Audit Sampling applies once the auditor has decided to test less than 100% of a population and draw conclusions about all of it. Issued by the IAASB, it has been effective for audits of financial statements for periods beginning on or after 15 December 2009. National versions follow it closely: NIA-ES 530 in Spain, NIA 530 in Latin America, ISA (UK) 530, SA 530 in India and, in the US, the AICPA's AU-C 530.
It works alongside ISA 500 Audit Evidence, which requires effective means of selecting items for testing: all items, specific items, or audit sampling. Only the last falls under ISA 530. Testing the ten largest invoices is selecting specific items, and the result says nothing about the rest of the population.
ISA 530 summary
| Section | Paragraphs | What it covers |
|---|---|---|
| Scope | 1-2 | Statistical and non-statistical sampling, tests of controls and of details |
| Effective date | 3 | Periods beginning on or after 15 December 2009 |
| Objective | 4 | A reasonable basis for conclusions about the population |
| Definitions | 5 | Ten terms, from sampling risk to tolerable misstatement |
| Sample design, size and selection | 6-8 | Purpose of the procedure, sufficient size, every unit can be selected |
| Performing audit procedures | 9-11 | Replacement items and items that cannot be tested |
| Nature and cause | 12-13 | Investigating deviations and misstatements; anomalies |
| Projecting misstatements | 14 | Projection in tests of details |
| Evaluating results | 15 | The sample result and whether sampling gave a reasonable basis |
| Appendix 1 | Stratification and value-weighted selection | |
| Appendices 2 and 3 | Factors driving sample size for controls and for details | |
| Appendix 4 | Sample selection methods |
Scope and objective
ISA 530 does not require you to sample. It applies when you have chosen to, with a statistical or non-statistical approach, for tests of controls and tests of details. Its single objective (paragraph 4) is that sampling gives you a reasonable basis for conclusions about the population the sample came from, which is why paragraph 15 asks not only what the sample showed but whether sampling gave you that basis.
Key definitions
- Audit sampling. Testing less than 100% of the items so that every sampling unit has a chance of selection.
- Sampling unit. A physical item (an invoice, a customer balance) or a monetary unit, as in monetary unit sampling.
- Sampling risk. The risk that your conclusion from the sample differs from the one you would reach testing everything. Wrongly concluding all is well affects effectiveness; wrongly concluding there is a problem affects efficiency. See sampling risk.
- Non-sampling risk. A wrong conclusion for other reasons: an unsuitable procedure, misreading evidence, missing a misstatement.
- Anomaly. A misstatement or deviation shown not to be representative of the population.
- Statistical sampling. Random selection plus probability theory to evaluate results and measure sampling risk. Without both, it is non-statistical.
- Tolerable misstatement. The amount you set to obtain appropriate assurance that actual misstatement does not exceed it, derived from performance materiality (ISA 320) and equal to or lower than it.
- Tolerable rate of deviation. The equivalent for tests of controls.
ISA 530 requirements
Design, size and selection (paragraphs 6 to 8)
Before selecting, define what counts as a misstatement for the purpose of the test. The application material uses receivables: a payment made before the confirmation date and received a few days later is not an existence misstatement. You also need evidence that the population is complete, usually by reconciling the listing to the ledger.
The sample must reduce sampling risk to an acceptably low level, and its size can come from a statistical formula or from judgement. Appendices 2 and 3 list the drivers: higher assessed risk, more assurance needed or more expected misstatement increase it; a larger tolerable misstatement, more evidence from other procedures or sound stratification reduce it. For large populations the number of items barely matters. The calculations are in audit sample size.
Every unit must have a chance of selection, and in statistical sampling a known probability. Appendix 4 describes random, systematic, monetary unit, haphazard and block selection, covered in audit sample selection methods.
Performing procedures (paragraphs 9 to 11)
If the procedure does not apply to an item (a properly voided cheque), you test a replacement. If you cannot perform it or a suitable alternative, for example because the documents are lost, the item counts as a deviation or misstatement. An unanswered confirmation is not yet one: subsequent receipts can be the alternative.
Nature and cause (paragraphs 12 and 13)
Each deviation or misstatement is investigated for its cause and its effect on the test and on other areas. If several share a feature (a transaction type, a location, a month), you may extend testing to every item with that feature, and they may point to fraud. Treating a misstatement as an anomaly is for extremely rare circumstances and needs a high degree of certainty, from additional procedures, that the rest of the population is unaffected.
Projection and evaluation (paragraphs 14 and 15)
For tests of details you always project misstatements to the population; for controls the sample deviation rate already is the projected rate. Anomalies can be left out of the projection but are still considered. When evaluating, if projected plus anomalous misstatement exceeds tolerable misstatement, the sample gives no reasonable basis; the closer it comes, the likelier actual misstatement exceeds it; and if projected misstatement exceeds the expected misstatement used to size the sample, sampling risk may be unacceptable. You can then ask management to investigate and adjust, or extend your procedures.
Audit sampling methods under ISA 530
The standard accepts statistical and non-statistical approaches and leaves the choice to judgement, adding that sample size is not a valid way to tell them apart. Common statistical methods are attribute sampling for controls and monetary unit or variables sampling for details (see attribute vs variable sampling). The full comparison is in statistical vs non-statistical sampling.
ISA 530, AU-C 530, AS 2315 and NIA-ES 530
| Standard | Issuer | Applies to | Relationship to ISA 530 |
|---|---|---|---|
| ISA 530 | IAASB | International baseline | The original text |
| AU-C 530 | AICPA | US audits of non-issuers | Clarified standard drafted from ISA 530 |
| AS 2315 | PCAOB | US audits of issuers | Separate standard (formerly AU 350) |
| NIA-ES 530 | ICAC | Statutory audits in Spain | Adapted translation with national notes |
| NIA 530 | National bodies | Colombia, Mexico, Peru and others | ISA 530 in Spanish |
AU-C 530 came in with the AICPA's clarified standards (SAS 122) and follows ISA 530's structure. PCAOB AS 2315 is older and uses its own terms: risk of incorrect acceptance and incorrect rejection for substantive tests, and assessing control risk too low or too high for controls. Its appendix links the risk of incorrect acceptance to the audit risk model. The AICPA Audit Guide: Audit Sampling is not a standard, but most methodologies take their tables from it, including the confidence factor of 3.00 at 95%.
ISA 530 worked example
Existence of a receivables balance of 4,812,350, monetary unit sampling, tolerable misstatement 120,000, expected misstatement 15,000, 95% confidence:
- Design (6). Population: debit balances reconciled to the ledger. Misstatement: a balance that does not exist or is overstated; cash in transit does not count.
- Size (7). 4,812,350 × 3.00 ÷ (120,000 − 15,000 × 1.6) = 150.4, rounded up to 151 items.
- Selection (8). Systematic over the cumulative listing, interval 4,812,350 ÷ 151 = 31,869.87, documented random start.
- Procedures (9 to 11). Confirmations, with subsequent receipts for non-replies. An item with no support would be a misstatement for its full amount.
- Cause (12). A balance of 12,400 should be 11,160 (10% tainting) and one of 3,050 should be 2,135 (30%). Both are late credit notes, so you test credit note cut-off.
- Projection (14). 0.10 × 31,869.87 + 0.30 × 31,869.87 = 12,747.95.
- Evaluation (15). Upper limit = 95,609.60 basic precision + 12,747.95 projected + 8,923.56 incremental allowance = 117,281.11. Below 120,000 but close, so you confirm the cut-off problem does not reach other items before concluding.
The upper limit is worked step by step in evaluating monetary unit sampling results.
Working paper checklist
- Objective of the test and assertion.
- Population: source, extraction date, reconciliation to the ledger.
- Sampling unit and definition of a misstatement or deviation.
- Items tested 100% or selected specifically, kept apart from the sample.
- Approach and method, with the reason.
- Parameters: confidence, tolerable misstatement linked to materiality, expected misstatement (or tolerable and expected deviation rates).
- Sample size calculation, reproducible.
- Seed or random start and the list of selected items.
- Result per item, replacements and untested items.
- Cause of each misstatement and evidence for any anomaly.
- Projection, comparison with tolerable misstatement and conclusion.
- Preparer and reviewer sign-off.
Frequently asked questions
What is ISA 530?
The International Standard on Auditing on audit sampling, issued by the IAASB. It governs how auditors design, size and select a sample, and how they project and evaluate the results, when they test less than 100% of a population.
What is the objective of ISA 530?
To provide a reasonable basis for the auditor to draw conclusions about the population from which the sample is selected (paragraph 4).
Does ISA 530 set a minimum sample size?
No. It requires a size sufficient to reduce sampling risk to an acceptably low level, set by a statistical formula or by professional judgement.
What is the difference between ISA 530 and AU-C 530?
AU-C 530 is the AICPA standard for US audits of non-issuers, drafted from ISA 530 in the clarity project with the same structure and requirements. Audits of US listed companies follow PCAOB AS 2315.
Which sampling methods does ISA 530 allow?
Statistical and non-statistical sampling. Appendix 4 describes random, systematic, monetary unit, haphazard and block selection. Haphazard selection is not appropriate for statistical sampling, and block selection rarely supports conclusions about a whole population.
Is selecting the largest items audit sampling?
No. Selecting specific items, such as everything above an amount, is a separate means of selection under ISA 500. Its results cannot be projected to the rest of the population.
- IAASB, ISA 530 Audit Sampling
- IAASB, ISA 500 Audit Evidence
- AICPA, AU-C 530 Audit Sampling
- PCAOB, AS 2315 Audit Sampling
- ICAC, NIA-ES 530 Muestreo de auditoría (Resolution of 11 April 2024)
- AICPA, Audit Guide: Audit Sampling
From the calculation to the working paper
AuditSampler Pro plans, selects and evaluates monetary unit, attribute and stratified samples under ISA 530, and produces the working paper as a PDF. Client data stays on your computer.