Cybersecurity
Retesting After a Penetration Test: Why Findings Must Be Verified
A fix that has not been tested is only a claim. A retest repeats the same attack to prove the flaw is closed. This guide explains the scope, the result statuses, and the preparation on your side.
SEO and Content Specialist
7 min read Reviewed by Legia Cybersecurity Team
Short answer: a retest is a follow-up test that repeats the same attack method against flaws that have been fixed, to prove that the flaw is truly closed and that the fix has not opened a new one. Without a retest, a penetration test report is a list of problems, not proof that the problems are resolved. The outcome is a status for each finding and a retest report that can be attached to an audit or tender.
After receiving a penetration test report, attention usually jumps straight to fixing. Many teams stop there: the ticket is closed, the code is deployed, and the finding is considered done. Yet “fixed” is the development team’s statement, while “closed” is a conclusion that has to be proven by trying the attack again. This article explains what a penetration test retest is, what it covers, the usual result statuses, and what to prepare before requesting one. How to read the original report is covered in How to Read a Penetration Testing Report.
Why fixes must be retested
NIST’s technical guide to information security testing and assessment (SP 800-115) says that the implementation of mitigation steps should be verified, among other ways by retesting the system and its components. The document also notes one condition: the test team can verify implementation only if a mirror copy of the original test is performed. In other words, a retest is not a new test with a different approach, but a repetition of the same steps to see whether the result is still the same.
The PCI Security Standards Council’s Penetration Testing Guidance (2017) is in line: after an organization remediates exploitable vulnerabilities, the tester performs a retest to validate that the new controls mitigate the original risk. That guidance is supplemental information and does not replace the requirements of the PCI standard, but the way of thinking applies generally.
There are three practical reasons for a retest:
- Fixes often close only the symptom. Restricting one input pattern, for example, is not the same as fixing how the application processes input.
- Changes can open new flaws. NIST notes that additional vulnerabilities may be uncovered in follow-up tests.
- Written evidence is useful beyond the technical team. A report containing the retest result can be attached as proof that testing was done and the flaws were closed.
Penetration test retest: what is retested
A retest repeats the testing for findings you state have been fixed, using the same method and accounts as the original test. The PCI guidance adds a point on scope: all changes resulting from remediation should be retested, while whether the whole system needs retesting depends on a risk assessment of those changes. If a fix touches a fundamental part of the application, the retest scope should be wider.
This distinction helps decide whether you need a retest or a new test:
| Retest | New penetration test | |
|---|---|---|
| Purpose | Confirm reported findings are closed | Look for flaws in a system that has changed |
| Scope | Fixed findings and changes resulting from the fixes | The whole agreed scope |
| When it fits | Fixes finish soon after the original test | Fixes take a long time or the system changed a lot |
| Result | A status for each finding | A report of new findings |
The PCI guidance notes that remediation that takes a long time after the original test may require a new test, because the state of the system has changed. The decision depends on a risk analysis of how much has changed since the original test.
Retest result statuses
The PCI guidance gives an example of retest report sections: executive summary, date of the original test, date of the retest, original findings, and retest results. For the “results”, the statuses below are commonly used. Status names vary between providers, so treat this as an example, not a standard:
| Status | Meaning | Follow-up |
|---|---|---|
| Closed | The same attack no longer succeeds | Record as done |
| Partially closed | The main path is closed, but variations still succeed | Fix the root cause, retest that part |
| Still open | The same attack still succeeds | The fix needs to be reviewed |
| New finding | The fix or another change opened a new flaw | Assess severity and schedule a fix |
Severity is read the same way as in the original report, for example through the score explained in What Is a CVSS Score. One thing to keep in mind: a penetration test proves flaws within the agreed scope and time. A “closed” status applies to that finding, not a guarantee that the system has no other flaws.
Preparing the fixes before requesting a retest
A retest goes quickly when the fixes are ready and documented. Prepare the following:
- Deploy the fix on the system to be tested. The PCI guidance stresses that when testing is done in a copy environment, flaws must still be fixed on production systems and the testing repeated. A fix in a test environment alone does not close the flaw your users are exposed to.
- Record the fix per finding: what was changed, where, when, and by whom. Refer to the finding number in the report.
- Fix the root cause, not only the example written in the report. If one page is vulnerable, check other pages that use the same pattern.
- Follow your company’s change process. NIST reminds that the wish for quick mitigation still needs to follow the organization’s change management policies and procedures.
- Keep the same test accounts and emergency contact if the original test used ordinary user accounts (grey box). This method is covered in Black Box, Grey Box, or White Box.
- Tell the tester that the fixes are done, along with the safest time window for testing.
Preparation for the original test, such as written authorization and an emergency contact, is covered in Preparing for Penetration Testing.
Fix patterns that often fail the retest
These patterns are common in remediation work in general, not findings from any particular client:
- The fix is only in the test environment. The code is correct there, but not deployed to production.
- Filtering is too narrow. Only the pattern written in the report is blocked, while variations still get through.
- Access checks only in the interface. The button is hidden, but the function can still be called directly.
- The fix does not touch every place. One endpoint is fixed, similar endpoints are left behind.
- The change brings new problems. A large fix alters the behavior of other parts of the system.
If a retest finds one or more of these patterns, it is not a project failure. It is the reason retests exist.
Keeping retest results for audits and tenders
A retest report that contains the original test date, the retest date, the original findings, and their statuses is stronger evidence than the original report alone. Keep both together with your remediation records. How to use reports for procurement is covered in Penetration Testing Reports for Tenders and Security Evidence for Client Audits or Tenders. Each auditor’s requirements differ, so match the report format to the requirements you received before testing begins.
How Legia does it
In the Penetration Testing service, retesting is the fifth stage and is included in the package. Legia re-tests the flaws that have been fixed, then issues a retest report and a closure statement for the flaws. This stage takes 2 to 4 working days and starts after you notify Legia that the fixes are done. The retest is included whoever fixes the flaws, your own team or Legia through Server Installation & Management or Website Maintenance, which are billed separately from the test. The whole process, from authorization to retest, generally takes 3 to 6 weeks, and the retest schedule follows when the fixes are finished on your side. Other security services are on the Cyber Security page.
Frequently asked questions
Do all findings have to be retested?
Findings you state have been fixed should be retested. The PCI guidance expects the retest report to cover all exploitable vulnerabilities that require remediation. Findings deliberately left unfixed should be recorded as accepted risk along with the reason, rather than left without explanation.
How long does a retest take?
At Legia, the retest stage takes 2 to 4 working days after notification that the fixes are done. Total project time depends more on how fast fixes happen on your side than on the testing itself.
What if the fixes are only finished months later?
The PCI guidance says drawn-out remediation can require a new test to get an accurate result for the current state. Share the expected fix schedule from the start, so the tester and you can decide whether a retest or a new test is the better fit.
Does a retest prove our system is secure?
No. A retest proves that the reported flaws have been closed. A penetration test, including its retest, covers only the agreed scope and time, and no test can guarantee that a system is free of other flaws.

