Audit and Assurance · Audit procedures

Computer-Assisted Audit Techniques (CAATs) and Automated Tools

Updated 11 October 2026 · Fact-checked

Computer-assisted audit techniques (CAATs) are auditing methods that use computers to test a client's data or systems. The main types are audit software, test data and data analytics. You solve exam questions by naming the technique, saying what it tests, and giving a benefit and a limitation that fit the scenario.

Understand Computer-Assisted Audit Techniques and Automated Tools

A CAAT is any use of a computer by the auditor to gather evidence. The client's records are often held in electronic form. Manual testing of thousands of items is slow, so the auditor lets software do the work.

There are two main families. Audit software runs on the client's data files. It can extract, sort, recalculate, compare and summarise data. Typical uses: recalculate depreciation, age receivables, find duplicate payments, select a sample, or list journals posted at weekends. This is mostly a substantive tool.

Test data works the other way round. The auditor feeds dummy transactions into the client's live or test system and checks that the system handles them correctly. For example, an invoice that exceeds a customer's credit limit should be rejected. This is mostly a test of controls tool, because it checks that programmed controls work. A related idea is an integrated test facility, where dummy records are processed with live data. Dummy items must be removed afterwards.

Data analytics uses software to analyse whole populations of data, often large volumes, to spot patterns, trends and outliers. Examples: compare sales by month and region, find unusual journal entries, or match purchase orders, goods received and invoices. Because it can test 100% of items, it can reduce the need for sampling. It also helps in risk assessment and fraud detection. The auditor still uses judgement to follow up the exceptions it finds.

CAATs are not a replacement for the auditor. The auditor must still check that the data is complete and reliable, and must understand the client's system. Results are only evidence if the data used is the right data.

Key rules to remember

Audit software: main uses: Extract → Recalculate → Compare → Analyse → Select samples; Mostly substantive procedures on the client's files, such as ageing receivables or finding duplicates.
Test data: purpose: Dummy valid and invalid transactions → client system → compare actual result with expected result; Tests programmed controls such as validation, limits and authorisation. Use both valid and invalid items.
Data analytics: key strength: Whole population tested, exceptions investigated; Can reduce sampling risk, but follow-up of exceptions needs auditor judgement.
Advantage and limitation pairing: Speed and coverage vs cost, set-up and data reliability; Give one point from each side, linked to the scenario.

How to solve Computer-Assisted Audit Techniques and Automated Tools questions

Use this method for any question on CAATs, whether it is an objective test item or a written requirement.

1Read the requirement. Decide if it asks for a technique, a use, advantages, limitations, or a recommendation.
2Identify the audit objective in the scenario, such as testing controls or testing balances and transactions.
3Match the tool: test data for controls in the system, audit software for substantive work on files, data analytics for large populations, trends and unusual items.
4Say exactly what the tool would do with the client's data in the scenario, for example recalculate, compare, or extract.
5Add a benefit that fits the facts, such as faster testing of a large volume of transactions.
6Add a limitation that fits, such as cost, need for IT skills, or the risk that the data is incomplete or inaccurate.
7Finish by saying how the result is used, such as follow-up of exceptions or a change in sample size.

Quickest way: Match tool to objective

When to use it: Use this for objective test questions and for short written points when time is tight.

Controls in a computer system? Choose test data.
Balances or transactions in large files? Choose audit software.
Whole population, patterns or unusual items? Choose data analytics.
Problem or drawback? Think cost, skills, data reliability, and risk of damaging client data.
Write one tool, one use, one benefit or limit, all tied to the scenario.

Common mistakes in Computer-Assisted Audit Techniques and Automated Tools

Saying test data tests the accuracy of balances in the financial statements.
Students mix up controls testing and substantive testing.
Fix: Test data checks whether system controls work. Audit software tests the data itself.
Listing only generic advantages like 'it is faster' with no link to the scenario.
Students learn lists and do not apply them.
Fix: Name the client's data, for example 'the receivables ledger of 8,000 accounts', and say what the tool would do.
Assuming CAATs remove the need for auditor judgement.
Data analytics output looks conclusive.
Fix: State that exceptions must be investigated, and that the auditor must assess the reliability of the data used.
Forgetting to say that data analytics can test a whole population.
Students treat it as just another sampling method.
Fix: Mention that it can analyse all items, which reduces sampling risk, but not that it guarantees no misstatement.
Ignoring practical risks, such as test data corrupting live records.
Students focus on benefits only.
Fix: Mention the need to remove dummy transactions, to run tests on a copy where possible, and to agree access with the client.

Worked examples

Example 1

An audit client has 12,000 receivables accounts held on a computerised ledger. Explain how the auditor could use audit software to obtain evidence, and give two limitations.

Show the solution

Identify the objective: test the valuation and existence of receivables, which is substantive work on the client's file.
Use of software: recalculate the ageing of each balance and compare to the client's ageing report.
Use of software: extract balances over credit limit or with no recent activity to focus on the risk of irrecoverable debts.
Use of software: select a sample of accounts for receivables confirmation.
Limitation 1: the software needs a copy of the file that is complete and agrees to the ledger totals, or the results are unreliable.
Limitation 2: set-up costs and IT skills are needed, which may not be worthwhile if the audit is small or the system changes often.

Answer: Audit software can recalculate ageing, extract overdue or over-limit accounts and select a confirmation sample from the 12,000 accounts. Limitations: it depends on complete and accurate client data, and it needs cost and IT skills to set up.

Example 2

A retailer's sales system is meant to reject orders from customers who have exceeded their credit limit. Explain how the auditor could test this control using test data and state one risk of doing so.

Show the solution

Identify the objective: a test of control over a programmed check in the sales system.
Prepare test transactions: some orders within the credit limit and some above it.
Predict the expected result: the first group is accepted and the second is rejected or flagged for authorisation.
Process the dummy orders through the system.
Compare actual output with expected output. Any order above the limit that is accepted shows the control is not working.
Risk: dummy transactions may corrupt live data or reach the ledgers, so the auditor should use a test copy or reverse the items and agree this with management.

Answer: Enter valid orders and orders above the credit limit, and compare the system's response with what was expected. If over-limit orders are accepted, the control is ineffective. Risk: dummy data could contaminate live records unless tested on a copy or removed.

Exam tips

Match the technique to the objective first. Examiners often reward the link between test data and controls, and audit software and substantive testing.
Always apply to the scenario. Name the data, the file or the control, not just the technique.
For advantages and disadvantages questions, give a balanced answer. One benefit and one limitation score better than three of the same type.
In objective test questions, watch for 'tests of controls' versus 'substantive procedures'. Wrong choices usually reverse these.
In written answers, say what the auditor does with the results, such as investigating exceptions or adjusting audit work.

Computer-Assisted Audit Techniques and Automated Tools in other exams

The same ground in other exams, if you are preparing for more than one or want another angle on it.

Computer-Assisted Audit Techniques and Automated Tools: frequently asked questions

What is the difference between test data and audit software?

Test data feeds dummy transactions into the client's system to see if controls work. Audit software runs on the client's existing data files to extract, recalculate or compare data. The first mainly tests controls and the second mainly gathers substantive evidence.

What are the advantages of CAATs in an audit?

They can test large volumes of data quickly and consistently. They can cover whole populations, find unusual items and reduce manual effort. They also let the auditor repeat tests easily in later periods once set up.

What are the disadvantages of CAATs?

They can be costly to set up and need IT skills. The results rely on the data being complete and accurate. There is also a risk of damaging client data, and some client systems may not be compatible.

How does data analytics help in an audit?

It lets the auditor analyse large sets of data for patterns, trends and outliers. It can support risk assessment, test whole populations and highlight fraud indicators such as unusual journals. The auditor must still investigate exceptions and judge their significance.