Skip to main content
QuantLab Logo

TOFU Education · 2026

Pentest vs Vulnerability Scan: What's the Difference?

They are not the same thing. They cost 10x apart. Compliance frameworks treat them differently. Here is the clean explanation of what each is, what each finds, and why your security program needs both.

By Bill Beltz, founder of QUANT LAB USA INC · Published May 12, 2026

Quick answer

A vulnerability scan is automated, costs $500 to $5,000 per quarter, and identifies known CVEs against an asset. A penetration test is human-driven, costs $8,000 to $40,000 per engagement, and validates exploitability by actually compromising the target. Scans answer "what is wrong"; pentests answer "what is an attacker actually able to do." Compliance frameworks like SOC 2, PCI DSS, and HIPAA require pentests, not scans.

Half the cybersecurity questions I get from founders boil down to this confusion. A vendor sold them a $4,000 quarterly Nessus scan and called it a pentest. The auditor flagged it. Now they are calling everyone in a panic.

They are different things. Both belong in a mature security program, neither replaces the other, and the distinction matters in writing more than verbally because compliance frameworks treat them differently. This is the clean explanation.

For deeper context, see what is penetration testing and our pentest service overview.

The clean comparison

AttributeVulnerability scanPenetration test
MechanismAutomated toolHuman-driven testing
GoalIdentify known CVEsValidate exploitability
Cost$500 to $5K per scan$8K to $40K per engagement
DurationHours to days1 to 3 weeks
FrequencyWeekly or continuousAnnual at minimum
CoverageBroad, shallowNarrow, deep
Finds business-logic flawsNoYes
Finds chained vulnerabilitiesNoYes
Finds zero-daysAlmost neverSometimes
Satisfies SOC 2No (supplement only)Yes
Satisfies PCI DSSQuarterly ASV part onlyAnnual requirement
OutputCVE list with CVSS scoresNarrative report with proof

What a vulnerability scan actually does

A vulnerability scanner is software that sends a curated set of probes against an asset and compares the response to a database of known CVEs (Common Vulnerabilities and Exposures). The scanner does not exploit, does not chain findings, and does not understand the business logic of the application. It says: this asset is running Apache 2.4.49 which has CVE-2021-41773, here is the CVSS score, patch it.

Modern scanners (Nessus, Qualys, Tenable.io, Rapid7) include thousands of plugins covering operating systems, web frameworks, network protocols, and cloud APIs. They are great at hygiene — finding the unpatched Wordpress install, the misconfigured S3 bucket, the open RDP port. They are useless at finding the auth-bypass logic flaw or the cross-tenant data exposure.

What a pentest actually does

A pentester is a human who tries to compromise the target the way an attacker would. They start with the scanner output as a baseline, then go beyond — they read the application, find logic flaws, chain multiple minor findings into severe compromises, and produce a written narrative of the attack path.

The typical web app pentest produces 15 to 40 findings, of which maybe 3 to 8 are critical or high. Most of the high-severity findings are things a scanner cannot detect: a parameter that lets you read another tenant's data, an auth flow that lets you bypass MFA, a session token that can be forged from a publicly visible identifier.

See our framework reference: what is penetration testing?, what is MITRE ATT&CK?, and what is OWASP Top 10?.

Mid-post: get the right test

Confused about whether you need a pentest or a scan? Free 30-minute call. We will tell you which is right for your compliance and threat model.

Why compliance frameworks treat them differently

Compliance frameworks codify the distinction because the failure modes are different:

  • SOC 2 (CC4.1, CC7.1): Requires "monitoring of controls" and "detection of vulnerabilities." Auditors interpret this as scanning plus pentest. Annual pentest minimum.
  • PCI DSS (11.3): Explicit. Quarterly external ASV scans (PCI 11.2) plus annual penetration testing (PCI 11.3). Both required.
  • HIPAA Security Rule: Requires risk analysis. Pentest is the implementation specification most auditors expect.
  • ISO 27001: Annex A.12.6.1 requires technical vulnerability management. Industry interpretation is scanning plus annual pentest.
  • NIST 800-53: Multiple control families (RA-5, CA-8) distinguish scanning from penetration testing.
  • FedRAMP: Annual penetration test mandatory by an approved third-party assessment organization (3PAO).

How they work together

The right pattern for most SaaS:

  1. Continuous: cloud configuration scanning (AWS Config, Wiz, Lacework). Daily.
  2. Weekly: vulnerability scan of all internet-facing assets (Tenable.io, Nessus).
  3. Monthly: vulnerability scan of internal infrastructure.
  4. Quarterly: vulnerability scan of the application surface (DAST tools like Burp Enterprise, OWASP ZAP).
  5. Annually: full penetration test against the application, API, and identity layer.
  6. Ad hoc: pentest after every major release that touches sensitive functionality.

This pattern catches both the daily flow of new CVEs (via scanning) and the deep architectural and business-logic flaws (via pentesting). Skipping either layer leaves a gap.

When a scan is enough

Honest cases where you do not need a pentest:

  • Pre-revenue, no compliance framework, no enterprise prospects, no sensitive data. Run scans, ship the product, defer the pentest to revenue.
  • Internal tools with no internet exposure and no sensitive data. Scans plus access control are enough.
  • Marketing websites without authentication or sensitive forms. Scans plus a CDN-level WAF are enough.

For everything else, you need both.

Tooling: what to use for scans

Honest 2026 recommendations:

  • Network and host scanning: Tenable.io or Nessus Pro. Tenable is the gold standard.
  • Web app DAST: Burp Suite Pro (manual-augmented) or OWASP ZAP (free).
  • Cloud configuration: Wiz, Lacework, or AWS Config + Security Hub for budget builds.
  • SCA (third-party dependency scanning): Snyk, Socket, GitHub Dependabot.
  • SAST (source code scanning): Semgrep, SonarQube, GitHub CodeQL.
  • Container scanning: Trivy (free), Snyk, Aqua.
  • PCI DSS ASV scanning: Trustwave, Qualys, ControlScan.

Frequently asked questions

What is the difference between a pentest and a vulnerability scan?

A vulnerability scan is automated and identifies known CVEs against an asset. A penetration test is human-driven and validates exploitability by actually attempting to compromise the asset. Scans cost $500 to $5,000 per quarter. Pentests cost $8,000 to $40,000 per engagement. Scans answer 'what is wrong'; pentests answer 'what is an attacker actually able to do.'

Is a vulnerability scan enough for SOC 2?

No. SOC 2 auditors universally interpret CC4.1 and CC7.1 as requiring annual penetration testing, not just vulnerability scanning. A scan-only program will be flagged. The right pattern: weekly automated scans for hygiene plus annual pentest for compliance evidence.

Is a vulnerability scan enough for PCI DSS?

PCI DSS explicitly requires both. Quarterly external ASV scans (Approved Scanning Vendor) plus annual penetration testing for in-scope environments. The two are not interchangeable — they cover different parts of the requirement.

How often should I run vulnerability scans?

Weekly for internet-facing assets. Monthly for internal infrastructure. Daily for cloud configuration. The goal is to catch new CVEs within the patch SLA, which for critical findings is 14 to 30 days under most compliance frameworks.

How often should I get a penetration test?

Annually at minimum. Quarterly for the highest-risk surfaces. After every major release that touches authentication, payment, or sensitive data. After every significant infrastructure change. SOC 2, HIPAA, and PCI all expect annual at minimum.

Can a pentest substitute for a vulnerability scan?

No. They cover different scope. A pentest covers a specific scope deeply at a point in time. A scan covers the full asset inventory continuously but shallowly. You need both. Pentest catches business-logic flaws and chained findings; scans catch the daily flow of new CVEs.

What does a vulnerability scan find that a pentest does not?

New CVEs that landed after the pentest ended. Scans run continuously and catch the next critical vulnerability the same week it is disclosed. Pentests are point-in-time. The 6-month-old pentest report does not tell you about the 30-day-old log4j-style vulnerability.

What does a pentest find that a vulnerability scan does not?

Business-logic flaws (the user can read another tenant's data via a parameter swap), authentication bypass chains (combining three minor flaws into a full compromise), and exploitation reliability. Scans cannot chain findings. Scans cannot detect missing access controls. Scans cannot exploit cryptographic weaknesses.

How much does a vulnerability scan cost?

Self-service tools like Nessus ($3K to $5K/year), Qualys ($5K to $30K/year), Rapid7 ($10K to $50K/year), or Tenable.io ($8K to $40K/year). Managed scan services run $5K to $25K/year. The biggest cost is interpreting the output — the scanner finds 500 things; you need a human to triage them.

Should I do scans in-house or outsource?

Modern engineering teams should run their own scans. The tools are easy. The output is the value. Outsourcing makes sense only for compliance scans (PCI DSS quarterly ASV scans must be performed by a certified vendor) and for environments where you do not have an in-house security engineer.

What does a clean vulnerability scan mean?

It means there are no known CVEs above your patch SLA threshold. It does not mean the system is secure. Most successful breaches exploit zero-days, business logic flaws, or chained vulnerabilities that no scanner detects. A clean scan is a baseline; a clean pentest is closer to evidence.

Can QUANT LAB USA help with both?

Yes. We run penetration testing engagements and advise on vulnerability scanning programs. We integrate with Nessus, Tenable.io, Qualys, and Rapid7. We help SaaS teams build the right cadence: weekly scans plus annual pentest, mapped to the compliance framework you are pursuing.

Pick the right test for the job.

Free 30-minute call. We will tell you whether your situation calls for a scan, a pentest, or both.

Or call Bill directly at (770) 652-1282
All blog postsUpdated May 12, 2026