Cybersecurity
How Often Does a Company Need a Vulnerability Assessment?
There is no single number that fits every company. This guide builds a vulnerability assessment schedule from asset type, how often systems change, and references such as CIS Controls and PCI DSS.
SEO and Content Specialist
5 min read Reviewed by Legia Cybersecurity Team
Short answer: for assets that are connected to the internet (websites, web applications, public servers), a monthly check is a sensible starting point. Internal assets are usually checked at least every three months. Beyond that schedule, run a check after every major change: a new module, a server migration, or a service newly opened to the public. CIS Controls v8 and PCI DSS v4.0 use similar frequency benchmarks.
A vulnerability assessment maps security flaws as widely as possible without exploiting them. The result is a snapshot of one point in time. The problem is that the snapshot goes stale quickly: new flaws are announced every week, components are updated, and configurations change. So the more useful question is not “have we ever checked”, but “how often should the check be repeated so the snapshot stays relevant”.
Why once a year is often not enough
Many companies run a security check once a year because that is what the auditor asks for. For audit purposes, an annual schedule may be adequate. For day-to-day security, the gap is too long, for three reasons:
- New flaws keep being announced. A component that is safe this month can have a flaw announced next month, and attackers usually start scanning for popular flaws soon after they are announced.
- Systems change more often than the audit schedule. Every release, new plugin, or configuration change can open a flaw.
- Fixes need to be confirmed. Without a recheck, there is no evidence that a reported flaw has actually been closed.
Frequency benchmarks from well-known standards
Two commonly used references state the frequency clearly:
| Reference | Internal assets | Internet-facing assets |
|---|---|---|
| CIS Controls v8, Safeguards 7.5 and 7.6 | Automated scans every three months or more often | Automated scans every month or more often |
| PCI DSS v4.0, Requirements 11.3.1 and 11.3.2 | Scans at least once every three months | Scans at least once every three months by an Approved Scanning Vendor (ASV) |
PCI DSS is mandatory only for organizations that process payment card data, but its numbers are useful as a comparison. CIS Controls is a set of security practices widely used as a general reference. PCI DSS also requires rescans to confirm that flaws have been closed, as well as scans after significant changes.
Building a schedule from asset type
A practical way to build a schedule is to group assets by how exposed they are and how often they change.
| Asset group | Examples | Sensible schedule |
|---|---|---|
| Internet-facing and changes often | A website with regular releases, a customer-facing web application, an API | Monthly, plus every major release |
| Internet-facing and rarely changes | A static profile page, a mail server | Monthly to every three months |
| Internal and important | A database server, internal applications, financial systems | Every three months, plus every major change |
| Internal and supporting | A file server, supporting devices | Every three to six months |
This table is a starting point, not a rule. Companies that store large amounts of personal data or handle online transactions usually choose a tighter schedule.
Triggers outside the schedule
A regular schedule sets the rhythm. Extra checks are still needed when:
- A new module or feature accepts input from users.
- A server is moved or the infrastructure changes, for example a migration to the cloud.
- A new service is opened to the public, such as an admin panel, an API, or remote access.
- A critical flaw is announced in a component you use.
- An incident occurs, to make sure no other flaws are left behind.
What makes regular checks actually useful
Frequent checks are not automatically safer if the results are not acted on. These four habits make the difference:
- The same assets, month after month. A fixed asset list makes trends readable: new findings, closed findings, and findings that keep recurring.
- Findings are verified. Raw scan results contain false positives. Without verification, the technical team tires quickly and starts ignoring the report.
- Clear priorities. Severity levels based on CVSS help set the order of fixes. The explanation is in the article What Is a CVSS Score?.
- Fix deadlines per level. Set a deadline for critical and high findings, then recheck after the deadline passes.
When a vulnerability assessment alone is not enough
A vulnerability assessment maps flaws, but it does not prove what could happen if a flaw were exploited. For systems that handle sensitive data, or before a major release, penetration testing complements it. A common pattern: regular vulnerability assessments throughout the year, plus penetration testing once a year or before a major change. The difference between the two is covered in the article Vulnerability Assessment or Penetration Testing?.
How Legia does it
In the Vulnerability Assessment service, Legia checks websites, applications, servers, and cloud environments. The check can be run once or repeated monthly on the same assets. Every finding is manually verified, given a severity level based on CVSS, and compiled into a prioritized report with remediation steps. Retesting after the fixes is included. One round of checks, from scoping to retest, usually takes 2 to 4 weeks.
Frequently asked questions
Is a weekly automated scan enough?
Frequent automated scans are useful for detecting changes, but they still need manual verification and risk assessment so the results can be acted on. Scanning without follow-up only produces an ever longer list.
What if our team has not had time to fix last month’s findings?
That is a sign the remediation schedule needs adjusting, not a sign the checks should stop. Prioritize critical and high findings, then agree on a realistic deadline for the rest.
Do assets in cloud services also need to be checked?
Yes. The cloud provider secures its infrastructure, but the configuration, operating systems, and applications you install remain your responsibility.

