Audit and Assurance · Automated tools and techniques

Test Data and Audit Software in ACCA Audit and Assurance

Updated 11 October 2026 · Fact-checked

Test data and audit software are computer-assisted audit techniques. Test data feeds dummy transactions into the client's system to see if programmed controls work. Audit software runs the auditor's own programs on the client's live data to extract, sort, recalculate and analyse it. Test data tests controls; audit software mainly tests data.

Understand Test Data and Audit Software

Many clients process transactions by computer. The auditor cannot always trace every step on paper. Computer-assisted audit techniques (CAATs) let the auditor use the computer itself to gather evidence. Two main types are test data and audit software.

Test data means the auditor enters made-up transactions into the client's system. Some are valid and some are deliberately wrong, such as a customer over its credit limit, an invalid employee number or a date in the wrong period. The auditor predicts what the system should do, then compares that with the actual result. If an invalid item is accepted, the control is not working. Test data is a test of controls. It tests how the program processes data, not the balances themselves.

Ordinary test data is often processed in a separate run, possibly on a copy of the program, rather than during live processing. So the auditor must confirm that the program tested is the same version the client uses in live processing. Otherwise the result says nothing about the controls that actually operate.

An integrated test facility (ITF) is a form of test data, but it is processed within live running. The auditor sets up a dummy entity, such as a fictitious customer or department, inside the live system. Test transactions are processed alongside real ones during normal processing. This can be done at any point or on a continuing basis. The dummy items must then be removed so they do not distort the real records. This gives evidence on controls operating during normal processing.

Audit software is a program the auditor uses on the client's real data files. Typical uses are: extracting items (such as invoices over a set value), recalculating (such as depreciation or receivables ageing), sorting and summarising, checking for duplicates or gaps in number sequences, comparing two files, and selecting samples. Audit software is mostly used for substantive procedures. The auditor can test 100% of the records in the file provided quickly.

Both need preparation. The auditor must understand the system, agree access with the client and make sure the data tested is the real data, not a copy that has been altered. Costs and skill needs are higher at the start, but repeat use in later years is cheap.

Key rules to remember

Test data: Test data = dummy valid and invalid transactions → expected result vs actual result; Used as a test of controls. Always predict the expected result before processing.
Integrated test facility: ITF = dummy entity in live system + test transactions processed with real data; Dummy entries must be reversed or removed afterwards.
Audit software: Audit software = auditor's programs run on client's live data files; Mainly substantive testing: extract, recalculate, sort, compare, sample, check duplicates and gaps.
Key distinction: Test data tests the program's controls; audit software tests the data; Use this one line to separate the two in any question.

How to solve Test Data and Audit Software questions

Use this method for any question asking you to explain, compare or apply test data, ITFs or audit software.

1Read the requirement. Decide if it asks for uses, advantages, disadvantages, or how to apply a technique to a scenario.
2Identify the purpose in the scenario. Is the auditor testing controls (test data or ITF) or testing balances and transactions (audit software)?
3Name the technique and define it in one sentence.
4Give specific uses or tests tied to the scenario, such as a receivables ageing recalculation or a payroll check for duplicate employee numbers.
5For test data, include both valid and invalid items and say you compare actual with expected results.
6Add practical points: client permission, understanding the system, removing dummy data, and testing the real live programs.
7If asked for pros and cons, give both and link each to the client's situation.
8Check each point is explained, not just listed, so it earns the mark.

Quickest way: Controls or data? Pick the tool in seconds

When to use it: Use in Section A objective test questions and when planning a written answer in Section B.

Ask: is it testing controls or data?
Controls: choose test data or ITF. Data: choose audit software.
Check the wording: 'dummy', 'fictitious', 'invalid transactions' point to test data; 'extract', 'recalculate', 'duplicates', 'gaps', 'whole population' point to audit software.
For ITF, look for a dummy entity inside the live system.
Eliminate options that say test data tests balances or that audit software tests input controls directly.

Common mistakes in Test Data and Audit Software

Saying test data and audit software are the same thing.
Both are CAATs and both involve computers.
Fix: Remember: test data tests controls in the program; audit software tests the client's data.
Listing only valid test transactions.
Students forget the aim is to see if controls reject errors.
Fix: Include invalid items such as an over-limit order or a missing field, and state the expected rejection.
Forgetting to remove dummy data after an ITF.
The focus is on the test, not the effect on live records.
Fix: State that test entries must be reversed or excluded so financial records are not misstated.
Giving generic CAAT answers with no link to the scenario.
Students memorise definitions only.
Fix: Name the actual data in the scenario, for example payroll or receivables, and say what the software does with it.
Claiming audit software removes the need for other procedures.
It can test 100% of the records in the file provided, which sounds complete.
Fix: Say the file provided is not necessarily the whole population of transactions. It gives evidence only on data the client holds. Completeness of the data and other assertions still need other evidence.
Ignoring that data must be reliable and real.
Students assume the file given is the live one.
Fix: Mention checking that the file used is the genuine live file, and testing it against the ledger totals.

Worked examples

Example 1

Section B style (part of a 20-mark question): Your audit client, a retailer, processes sales invoices and credit control automatically. The audit manager proposes using test data. Explain how test data would be used and one limitation. (6 marks)

Show the solution

Define: test data means entering dummy transactions into the client's system to check that programmed controls work.
Design: prepare valid and invalid transactions, for example an order within a customer's credit limit, one above the limit, and an invoice with a missing customer code.
Predict: before processing, decide what the system should do. The over-limit order should be blocked or flagged, and the missing code should be rejected.
Process and compare: run the items through the system and compare actual output with expected results. Any invalid item accepted shows a control weakness.
Practical point: process on the live system, or ensure the program tested is the same as the one used in live processing. Reverse any test entries so records are not distorted.
Limitation: it tests only the controls and conditions the auditor thought of, and it gives no evidence on actual balances. It also needs good system knowledge, which takes time.

Answer: Enter valid and invalid dummy transactions, predict the expected system response, process them and compare results. Differences show control failures. Limitation: it tests only the conditions designed, is time-consuming to prepare, and does not directly test balances.

Example 2

Objective test: An auditor wants to check whether the client's payroll file contains duplicate employee bank account numbers and to recalculate gross pay for all employees. Which technique is most suitable? A Test data B Audit software C Integrated test facility D Inspection of a sample of payslips

Show the solution

Identify the purpose: checking real payroll data for duplicates and recalculating pay across all employees.
This is testing data, a substantive procedure, not testing whether a control rejects invalid input.
Test data and ITF use dummy transactions to test controls, so A and C do not fit.
Inspecting a sample covers only some employees and is manual, so D does not cover the whole file.
Audit software can run on the full live file, find duplicates and recalculate pay.

Answer: B Audit software

Exam tips

Always state the purpose: test data and ITF test controls; audit software performs substantive tests on data.
In written answers, give scenario-specific examples, such as aged receivables recalculation or duplicate supplier payments.
For advantages, use points like whole population tested, speed and reuse in later years. For disadvantages, use set-up cost, skills needed, risk of corrupting live data and reliance on the right file being used.
In objective questions, watch the keywords: dummy and invalid items mean test data; extract, recalculate and duplicates mean audit software.
Do not forget to compare actual with expected results when describing test data.

Test Data and Audit Software: frequently asked questions

What is the difference between test data and audit software?

Test data sends dummy transactions through the client's system to see if the programmed controls work. Audit software runs the auditor's programs on the client's real data to extract, recalculate, sort or compare it. One tests controls, the other tests data.

What is an integrated test facility in audit?

It is a dummy entity, such as a fictitious customer or cost centre, set up within the client's live system. Test transactions are processed with real ones during normal operation. The dummy entries must be removed afterwards so they do not affect the real records.

What are examples of audit software uses in ACCA AA?

Common examples are recalculating depreciation, ageing receivables, listing items over a set value, finding duplicate payments, identifying gaps in invoice number sequences, comparing files year on year and selecting samples. Always link the use to the data in the scenario.

Is audit software a test of controls or a substantive test?

It is mainly used for substantive procedures because it works on the client's data. It can give some indirect evidence on controls, for example by spotting gaps or duplicates, but its main role is testing transactions and balances.