Key takeaways
- Banks are required to test their key security controls regularly. The requirement comes from the interagency guidelines that implement GLBA.
- No rule sets a fixed schedule. The institution’s own risk assessment sets how often and how deeply to test.
- Tests must be run or reviewed by people independent of those who build and maintain the security program.
- Examiners read the report, the board minutes, and the remediation tracking together.
Where the requirement comes from
The Interagency Guidelines Establishing Information Security Standards implement section 501(b) of GLBA for banks. They direct each institution to regularly test the key controls, systems, and procedures of its information security program.
The guidelines add two conditions. The frequency and nature of the tests are determined by the institution’s risk assessment. And the tests are conducted, or their results reviewed, by independent third parties or by staff independent of those who develop or maintain the program.
Credit unions follow a parallel rule from the NCUA.
What the FFIEC handbook adds
Examiners work from the FFIEC IT Examination Handbook. Its Information Security booklet treats penetration tests, vulnerability assessments, and audits as separate forms of assurance that answer different questions.
- A vulnerability assessment identifies known weaknesses across systems.
- A penetration test shows whether an attacker can use weaknesses to reach systems or data, and how far they get.
- An audit checks whether the program follows policy and the guidelines.
One does not replace another. An examiner who asks for a penetration test and receives a scan report will note the gap.
The Cybersecurity Assessment Tool is retired
Many institutions used the FFIEC Cybersecurity Assessment Tool to measure their own maturity and to explain their testing schedule. The FFIEC retired it on August 31, 2025.
The FFIEC pointed institutions to current resources instead, including the NIST Cybersecurity Framework 2.0 and the CISA Cybersecurity Performance Goals, along with resources developed by the industry. If your testing schedule still cites the retired tool, move the reasoning to the framework you use now.
What a bank penetration test should cover
Scope follows the risk assessment. For most institutions that means the following.
- External network. Everything reachable from the internet, including remote access and email.
- Internal network. What an attacker can do from inside, such as from a compromised workstation in a branch. This includes Active Directory, privileged accounts, and the paths toward core banking and wire systems.
- Online and mobile banking. Authentication, session handling, and whether one customer can reach another customer’s accounts.
- Social engineering. Phishing and phone pretexting against staff, where the risk assessment calls for it.
- Segmentation. Whether the controls that separate sensitive systems, such as the cardholder data environment, actually hold.
Much of this depends on vendors. Testing a hosted core or a third party banking platform needs the provider’s permission and is often limited to your own configuration and integration. Settle that before the test starts.
What examiners look for in the result
The report is one part of what an examiner reads. They also look for evidence of what the institution did with it.
- Independence. Who performed the test, and whether they are separate from the team that runs the systems.
- Scope tied to risk. Why these systems were tested, with a line back to the risk assessment.
- Findings ranked by real exposure. Severity based on what an attacker could reach in your environment, not only a stock CVSS score.
- Tracking to closure. Each finding assigned to an owner and followed until it is fixed or formally accepted.
- Reporting to the board. Results and remediation status presented to the board or its committee.
A retest helps with the fourth point. It adds an addendum to the report that shows which findings were tested again and what was observed.
How ForgePath helps
We run external, internal, application, and social engineering tests as an independent firm, and we scope each one to your risk assessment. Reports rank findings by what an attacker could reach in your environment, show the evidence of how we exploited each one, and include a summary written for the board. Retesting is available as an add on.
See how we work with banks, credit unions, and fintech companies, or read about FFIEC readiness, external network penetration testing, and internal network penetration testing.
Frequently asked questions
Does a bank need a penetration test every year?
No rule sets that schedule. The guidelines require regular testing at a frequency your risk assessment supports. Whatever schedule you choose, document the reasoning.
Can our internal IT team perform the test?
Only if the testers are independent of the people who develop or maintain the security program. For most community institutions that means an outside firm.
Does a penetration test satisfy the IT audit?
No. An audit and a penetration test answer different questions, and examiners expect both.

