Test Log in Software Testing: Structure, Contents and Example
A test is only worth something if you can later trace exactly what was checked and with what result. That is what the test log is for. It is the evidence that testing took place, the basis for defect reports and, in the end, the foundation for the release decision. This article shows what belongs in a test log, what an example looks like and how it differs from a test report.
What is a test log? Definition
The ISTQB glossary describes the test log as a chronological record of relevant details about the execution of tests. In practice this means: while a tester executes a test case, they record when they did what, with which software version and in which environment, which result they expected and which result they actually got.
The terms are not always used consistently. Some teams speak of test records, test evidence or test run logs and mean the same thing. What matters is the core idea: the test log documents execution, not planning. What is to be tested is described in the test case and the test plan. What actually happened during testing is recorded in the test log.
What is a test log used for?
- Traceability: even weeks later it is clear which version was tested with which data.
- Defect analysis: developers see the conditions under which a defect occurred and can reproduce it faster.
- Evidence for acceptance: the client, business department or auditors receive reliable records instead of the statement "it was tested".
- Comparing test runs: if a test case passed two releases ago and fails now, this points to a regression.
- Basis for the test report: test management summarises many test logs into a status for project leadership and management.
What belongs in a test log?
A proven structure is a header with the general conditions and a results section per test case or test step.
Header: conditions of the test run
- Test object and version: application, build or version number. Without the exact version a log is hardly usable.
- Test environment: for example test system, browser, operating system or device. More on this in our article on building a test environment.
- Test run or test cycle: such as "Regression test release 4.2" or "Acceptance test round 1".
- Date and time of execution.
- Tester: who executed the test.
Results section: per test case and test step
- Test case ID and title as a reference to the planned test case.
- Test data used: concrete values, for example a test customer or an order number.
- Expected result according to the test case.
- Actual result: what really happened, as concretely as possible.
- Status: passed, failed, blocked or not executed.
- Defect ID: reference to the ticket if a deviation was reported.
- Evidence: screenshots, log excerpts or files where they are needed for the assessment.
- Remarks: observations that are not defects but worth mentioning.
How the underlying test case is structured is described in the article how to write test cases.
Test log example: a test case from practice
The following example shows a lean test log for a manual functional test in an online shop. The values are a fictitious example.
Header
- Test object: online shop, version 4.2.0 (build 118)
- Environment: test system "Staging", Chrome on Windows 11
- Test run: regression test release 4.2
- Date / tester: 12 March, 10:15 / M. Example
Test case TC-031: voucher is applied in the shopping cart
- Test data: test customer "customer-07", item worth €80.00, voucher code "TEST10" (€10 discount)
- Step 1: add item to cart. Expected: cart shows €80.00. Actual: €80.00. Passed.
- Step 2: enter and confirm voucher code. Expected: message "Voucher applied", total €70.00. Actual: message appears, total stays €80.00. Failed, defect BUG-214, screenshot attached.
- Step 3: proceed to checkout. Blocked by BUG-214.
- Overall status TC-031: failed
The example shows two typical points: the actual result is described concretely, not just as "doesn't work". And steps that cannot sensibly be executed because of a defect are explicitly marked as blocked instead of simply being left out.
Test log vs. test report: the difference
The two terms are often confused. The test log records execution in full detail and without gaps, usually per test case. It is aimed at testers and developers. The test report summarises the results of many test logs: how many test cases were executed and passed, which defects are open, which risks remain, and can a release be recommended from a testing perspective? It is aimed at project leadership, the business and management and is usually prepared by test management.
In short: the test log answers "What happened in this test?", the test report answers "What is the state of quality and can we go live?".
Test logs for manual and automated tests
With automated tests, test tools and CI pipelines usually generate the logs themselves. Here the challenge is to present the results in a readable way and store them permanently. With manual testing, on the other hand, the log is created by the tester, ideally directly during execution in a test management tool, or alternatively in a structured spreadsheet.
For exploratory tests without pre-written test cases, session notes are sufficient: the goal of the test session, areas covered, duration, defects found and open questions. That is a test log too, just in a freer form.
Tips for meaningful test logs
- Record immediately: anyone who documents from memory at the end of the day loses details.
- State version and environment exactly: "current version" is worthless after the next deployment.
- Blocked is not failed: if a test could not be executed because of another defect or because an interface was down, that is not a defect in the function under test. The distinction keeps the test status honest.
- Use evidence selectively: a screenshot for every deviation helps; a screenshot for every passed step usually just bloats the log, unless acceptance requires it.
- Respect data protection: screenshots and logs should not contain real personal data. If this cannot be avoided, clarify the processing with the client and data protection in advance. More in the article test data management.
- Store with versions: logs from earlier test runs are valuable, for example when comparing regression tests across several releases.
Test logs in acceptance testing
Test logs become particularly important in acceptance testing. This is where the client decides whether the software meets the agreed acceptance criteria. Complete logs make that decision traceable and help later if questions about warranty or rework arise. It is therefore worth agreeing in the test plan which details and evidence the logs should contain.
If your project lacks the time or people for clean test documentation, an external testing partner can help. As a software testing service provider in Hamburg we carry out manual software tests, document test execution in a traceable way and deliver a test report with a release recommendation at the end, remotely, hybrid or on site in Hamburg and across Germany.
Conclusion
A good test log is not bureaucracy for its own sake. It makes tests traceable, speeds up defect fixing and provides the evidence an acceptance decision can rely on. With a clear header, concrete actual results, an honest status per test case and targeted evidence, a lean template is often enough.
Frequently asked questions about test logs
What is the difference between a test log and a test report?
The test log documents the execution of individual tests in detail: version, environment, steps, expected and actual results and the status. The test report summarises the results of many test logs, shows open defects and risks and usually contains a release recommendation.
Who creates the test log?
The test log is created by the person who executes the test, usually the tester. With automated tests the test tool generates the log. The test report based on the logs is usually prepared by test management.
Is there a template for a test log?
There is no mandatory template. To get started, a spreadsheet with the columns test case ID, test step, expected result, actual result, status, defect ID and remarks is enough, plus a header with test object, version, environment, date and tester. In larger projects a test management tool provides this structure.
Traceable software testing for your project
We test your software manually and in a structured way, document test execution in a traceable way and deliver a test report with a release recommendation from a testing perspective.
Book an initial consultationMore about our service: Software testing & quality assurance in Hamburg →