A fixed sample size like "25 for any population over 500 items" is a firm habit, not a statistical result — the actual drivers are the confidence level wanted, the tolerable error or deviation rate, and the error or deviation rate already expected, and population size barely moves the answer once a population is reasonably large. Two populations of 2,000 and 20,000 items, tested at the same confidence and tolerable error, produce almost the same sample size; what actually changes the number is tightening the tolerable error or raising the confidence level.
What actually drives the number
| Input | Effect of tightening it | Effect of loosening it |
|---|---|---|
| Confidence level (90% → 99%) | Sample size roughly doubles | Sample size roughly halves |
| Tolerable error (relative to population value) | Smaller tolerable error inflates the sample sharply | Larger tolerable error shrinks it sharply |
| Expected error rate | Any expected error inflates the required sample above the basic calculation | Zero expected error gives the smallest sample the other inputs allow |
| Population size (once reasonably large) | Negligible effect past a few thousand items | Negligible effect past a few thousand items |
Treating population size as the main driver is the most common sampling mistake in practice — it feels intuitive ("bigger population, bigger sample") but for populations past a few thousand items, it's close to the least important input in the calculation.
Why a firm-wide fixed number persists anyway
A standard sample size is easy to apply consistently across files and easy for a reviewer to recognise — which is exactly why many firms default to one. The cost shows up on the populations where it's wrong: a low-risk, low-tolerable-error-sensitivity population gets over-tested, and a population where the real tolerable error is tight gets under-tested, with nobody noticing because the number matched what every other working paper in the file used.
Running the actual calculation — confidence level, tolerable error or deviation rate, expected error rate — takes the same five minutes a fixed-number lookup does, and produces a number that's actually defensible against the specific population being tested, rather than against "what we always use."
This is practical guidance written for audit teams, not an authoritative statement of any standard or statute. Standards, forms and timelines change. Check the current text issued by the ICAI and the relevant authority before relying on anything here in an audit file.
Related resources
- Audit workflow & automationSA 530: what the audit sampling standard actually requiresSA 530 doesn't mandate statistical sampling. What it mandates is that every sample, statistical or not, has to let you project results to the whole population.
- Audit workflow & automationAudit sampling methods for Indian CA firmsStatistical vs judgmental sampling, when full-population testing removes the need to sample, when sampling still applies, and what drives sample size.
- Purchase & AP auditPurchase testing: a complete exception taxonomyEvery category of finding a three-way match can produce, what each usually means in an Indian mid-market client, and which are worth escalating versus documenting and moving on.
Free tools, no signup
Each one reads a file entirely in your browser — or, for two of them, just answers a few questions — and nothing is ever uploaded.
- Purchase register checkDuplicate payments, round-sum amounts, weekend-dated invoices, incomplete records.
- GSTR-2B reconcilerMatches your purchase register against GSTR-2B — missing invoices either side, amount mismatches.
- Bank reconciliation checkerMatches your cash book against a bank statement — what's outstanding on each side.
- Fixed asset register checkDuplicate asset codes, missing location or useful life, negative net book value.
- Payroll red-flag scannerBank accounts shared across employees, and records missing an ID, PAN or salary.
- CARO applicability checkerAnswer a few questions about the company and see which CARO 2020 clauses likely apply.
- MSME payment delay checkerMSME-flagged invoices paid — or still unpaid — beyond the 45-day Section 22 limit.
- Related-party transaction screenerTransactions above a threshold you set with no approval reference, grouped by counterparty.
- Depreciation useful life checkerCompares a useful life in use against the Schedule II, Part C indicative figure.
- Materiality calculatorOverall and performance materiality against PBT, revenue, assets or equity, with reasoning shown.