Join the waitlist

Random number generator for audit sampling.

How many items do I need to look at, and which ones?

Pick item numbers at random from a population range, with no duplicates and a seed you can write in the workpaper so a reviewer can repeat the selection.

  • Runs in your browser
  • Nothing stored or sent
  • Free, no sign-up
Population and sample
The same seed, range and sample size always give the same items. Record the seed in the workpaper so a reviewer can re-perform the selection.

Numbers outside the range you enter are never picked. If the population has gaps, such as voided documents, draw extra items and skip the gaps in order, and say so in the file.

Items selected
60
No duplicates
Population
5,000
1 to 5000
Selection rate
1.2%
Items selected ÷ population
Selected item numbers (sorted)
16, 18, 30, 150, 164, 190, 223, 290, 430, 487, 493, 555, 591, 653, 789, 814, 890, 905, 1196, 1237, 1259, 1387, 1644, 1909, 2076, 2232, 2238, 2264, 2265, 2315, 2459, 2617, 2627, 2660, 2677, 2872, 3019, 3057, 3072, 3082, 3221, 3226, 3251, 3276, 3338, 3461, 3492, 3531, 3623, 3680, 3809, 4044, 4204, 4248, 4359, 4411, 4432, 4525, 4645, 4701
Method documented
Work shownrecomputed as you type
Population = 5000 − 1 + 1 = 5,000 items
Seed = "FY2026-REVENUE" → 32-bit hash → mulberry32 generator
Draw 60 of 5,000 without replacement (partial Fisher-Yates), then sort

This is a simple random selection: every item has the same chance of being picked. Random selection is one of several selection methods AU-C 530 recognizes (others include systematic, monetary unit and haphazard selection), and statistical sampling requires a random-based method. The generator is a repeatable pseudo-random algorithm, which suits audit documentation because it can be re-performed; it is not a cryptographic source. Selecting at random removes selection bias but does not set the sample size, which should come from your sampling plan.

Random selection in an audit sample

AU-C 530 has the auditor select items so that each sampling unit in the population has a chance of selection, and the sample can reasonably be expected to be representative. Random selection is one common way to do that, alongside systematic selection, monetary unit sampling, haphazard selection and block selection. With random selection every item has the same chance of being picked, so the auditor's own preferences cannot bias the sample. Statistical sampling requires a random-based method (random number selection, or systematic or monetary unit selection with a random start); haphazard selection is acceptable only in nonstatistical sampling and is easy to bias.

To use the tool, number every item in the population from first to last (invoice numbers, row numbers in a listing) and enter that range. The tool draws without replacement, sorts the result, and builds a one-line method note you can paste into the workpaper. It decides which items to test, not how many: take the sample size from your sampling plan, such as the attribute sample size or monetary unit sampling calculators.

Common questions

How do I select random numbers for an audit sample?

Number each item in the population, enter the first and last number and the sample size, and the tool returns that many unique item numbers in ascending order. Look up each number in the population listing and test that item. Record the range, the sample size and the seed.

What is the seed and why record it?

The seed is the text that starts the pseudo-random generator. The same seed, range and sample size always return the same items, so a reviewer can re-perform the selection and get your sample. Change the seed and you get a different, equally valid sample.

Is haphazard selection the same as random selection?

No. Haphazard selection means picking items without a structured technique and trying to avoid bias. AU-C 530 recognizes it as a selection method for nonstatistical sampling, but it is hard to show it is free of bias, and it is not appropriate for statistical sampling, which needs a random-based method.

Why are there no duplicates?

The tool samples without replacement: an item that has been chosen is not eligible again. Item-based audit samples are usually drawn this way because testing the same item twice adds no evidence.

Can I use this for monetary unit sampling?

Monetary unit sampling selects dollars, not items, usually with a fixed interval and a random start. Use the monetary unit sampling calculator for that. This tool is for item-based selection, such as attribute samples for tests of controls.

Runs in your browser. The figures you type are not sent to CORAA or stored. Check the result against your firm's methodology before relying on it.

Related audit calculators

CalculatorAudit Sample Size Calculator: Attribute Sampling (AU-C 530)Sample size for tests of controls from confidence, tolerable and expected deviation rates, plus the achieved upper deviation limit.CalculatorStratified sample selector by dollar threshold (key items + random)Test every item above a threshold and pick the rest at random by stratum, with item counts and dollar coverage shown.CalculatorMonetary Unit Sampling Calculator (MUS / PPS)Reliability factor, sampling interval and sample size, then the upper misstatement limit from the errors you find.
For US CPA firms

Join the CORAA waitlist

CORAA AI drafts audit working papers from your client’s QuickBooks books and does the first pass: lead schedules, tie-out, memos, draft notes and letters. CORAA Network reviewers clear every finding. Your firm reviews and signs.

  • First audit free, so you can judge the quality on a real file
  • Flat fee by entity size; messy books included
  • We work only with CPA firms, never their clients
See how it works →

* required