Skip to content

Stockertim Dictionary · Audit and assurance

Analytical procedures

Also called analytical review, substantive analytical procedures, analytics

In short

Evaluations of financial information by studying plausible relationships among financial and non-financial data, and investigating fluctuations that do not fit.

  • Intermediate
  • Process
  • Checked

In detail

Analytical procedures compare what the numbers are with what the auditor expects them to be. The expectation might come from last year's figures, budgets, industry data, or relationships such as payroll cost against headcount, interest against average debt, or revenue against units sold. Where the recorded figure differs from the expectation by more than a set threshold, the auditor investigates.

They are used at three stages. At planning, they help identify risks, such as a margin that has jumped without obvious reason. As substantive procedures, they can provide evidence on their own for predictable balances, provided the expectation is precise enough and built from reliable data. At completion, they give an overall check that the financial statements make sense in light of what the auditor has learned.

The strength of an analytical procedure depends on the quality of the expectation. A broad comparison with last year gives little assurance; a model that rebuilds rental income from the lease register and occupancy data can give a great deal. Explanations for differences must be corroborated with evidence, not just accepted from management.

Worked example

A company has 250 staff paid an average of 3,200 a month. The ledger shows annual salaries of 10,100,000.

  1. Expected salaries: 250 × 3,200 × 12 = 9,600,000
  2. Difference: 10,100,000 - 9,600,000 = 500,000
  3. Difference as a share of expectation: 500,000 / 9,600,000 = 5.2%

If the threshold is 3%, the 5.2% difference must be investigated and explained with evidence.

In practice

Auditors build expectations in spreadsheets or analytics tools, often across whole populations of transactions. Planning analytics are a standard part of the risk assessment file, and final analytics are reviewed by the engagement partner.

Watch out

  • A difference explained by management is not resolved until the explanation is corroborated.
  • Analytics on unreliable data give false comfort.

Learn it properly

Official sources

Facts last checked on 3 November 2026.

Education, not advice: the Stockertim Dictionary explains what terms mean. It never recommends buying or selling anything.

English