Find the way in from the internet before an attacker does.
InfoSight's U.S.-only, certified testers attack your public IP ranges, firewalls, VPN gateways, remote-access portals and exposed services from the internet, inside a scope you approve up front. Every finding is confirmed by hand and ranked by business impact.
Where the test starts
The internet. An external test typically runs black box, with no accounts and no inside knowledge, unless the scope says otherwise.
What sets the scope
Live external IP addresses, CIDR blocks and hostnames
Rules that require testing from outside
PCI DSS Requirement 11.4.3 and NYDFS 23 NYCRR 500.5(a)(1), which also requires testing from inside. The other rules below do not require it; the FTC rule's definition allows testing from outside or inside.
NIST's testing guide, SP 800-115, defines external security testing as security testing conducted from outside the organization's security perimeter: a view of your security posture as it appears from outside, usually from the internet, aimed at what an external attacker could exploit.
In NIST's definition of penetration testing, evaluators mimic real-world attacks to find ways to circumvent the security features of an application, system or network, and most tests look for combinations of vulnerabilities that give more access than any single one could.
A vulnerability assessment answers a different question: NIST's glossary describes it as a systematic examination of a system to determine the adequacy of security measures and identify security deficiencies.
The approved internet-facing systems in scope, typically tested from an unauthenticated internet perspective, unless authenticated testing is agreed in the scope. A typical scope includes:
Every live address and DNS name in scope, including hosts that public registration and DNS records reveal, and the services listening on each.
The devices that decide what gets in. Testers look for rules that allow too much, exposed administration and the unpatched gateway that opens the door.
VPN logins, remote desktop gateways and any other portal that lets staff reach internal systems from outside.
Mail gateways, DNS servers and web servers at the network layer: operating system, encryption settings and open ports; the application behind the login is typically scoped as a web application test.
File transfer, remote administration, database ports and device web consoles: anything reachable from the internet, whether or not it was meant to be.
Public load balancers, gateways and virtual machines you run in AWS, Azure or Google Cloud, tested within each provider's published testing rules.
InfoSight's baseline methodology runs in five phases. On an external test they look like this.
The limits go in writing first, in what NIST calls the rules of engagement: guidelines and constraints for the test, established before it starts. Yours name the targets, the testing window, a contact who can pause testing and anything off limits. NIST's guide also notes that excluding tests known to create denial-of-service conditions helps reduce disruption; any disruptive test runs only where the rules of engagement expressly authorize it and every provider involved permits it.
For network testing, InfoSight scopes by the number of live external IP addresses. Every address in scope has to be yours, or covered by written authorization from the system's owner; hosting and cloud providers' rules apply as well. NIST's guide says assessors should verify that all public addresses on a target list are under the organization's purview before testing begins.
The ranges security vendors publish put an external network penetration test at about $5,000 to $20,000, rising with the number of live IP addresses. Those are published market ranges, not InfoSight prices or a quote (see what a penetration test costs).
Book a zero-cost 15-minute scoping call and instantly receive a preliminary scan report to share with stakeholders and executives.
Prefer the phone? Call 305.828.1003.
They often sit side by side in one proposal, but they answer different questions.
| Service | Starts from | The question it answers | How it works |
|---|---|---|---|
| External penetration test | The internet, usually with no accounts | Can an outside attacker get in, and how far? | Testers exploit and chain weaknesses in internet-facing systems; every finding is confirmed by hand |
| Vulnerability scan | Wherever the scanner runs, inside or outside | Which known weaknesses are present? | Automated checks identify hosts and their known vulnerabilities; the work generally stops at identification rather than manual exploitation and chaining |
| Internal penetration test | A foothold already inside: a phished laptop, a plugged-in device, a contractor's account | How far can an intruder move once inside? | Testers escalate privileges and move laterally from inside, toward your most sensitive data |
If you plan both, NIST's guide notes the external test usually comes first, so testers carry no inside knowledge into it. Between tests, InfoSight runs vulnerability assessments as a separate service.
Which of these apply depends on the data you hold and who regulates you. Of the rules below, only PCI DSS 11.4.3 and NYDFS 500.5(a)(1) require testing from outside; the FTC rule's definition allows either perspective, and the others do not say where a test runs from.
Entities in PCI DSS scope must run external penetration tests (11.4.3) at least once every 12 months and after any significant infrastructure or application upgrade or change, using a qualified tester who is organizationally independent. Exploitable findings must be corrected and the test repeated to verify the corrections (11.4.4). Where segmentation isolates the cardholder data environment, segmentation controls are tested at least every 12 months and after changes to them (11.4.5); service providers test them at least every six months and after changes (11.4.6). PCI DSS is a payment-industry standard enforced through card brands and acquirers, not a statute.
Financial institutions under the FTC's jurisdiction must conduct annual penetration testing, scoped each year by their risk assessment, plus vulnerability assessments at least every six months, after material changes, and whenever there are circumstances they know or have reason to know may have a material impact on their security program, unless they run effective continuous monitoring or other systems that detect vulnerability-creating changes on an ongoing basis. The rule's definition of penetration testing covers attempts to penetrate databases or controls from outside or inside your information systems. Institutions holding customer information on fewer than 5,000 consumers are excepted from that testing requirement (16 CFR 314.6).
Entities that operate under, or are required to operate under, a license, registration, charter, certificate, permit, accreditation or similar authorization under New York's Banking, Insurance or Financial Services Law must, at a minimum, have a qualified internal or external party conduct penetration testing of their information systems from both inside and outside the systems' boundaries at least annually. Smaller covered entities that meet one of the size tests in 23 NYCRR 500.19(a) are exempt.
The bank regulators' Interagency Guidelines and NCUA's guidelines for credit unions call for regularly testing the key controls, systems and procedures of the security program, at a frequency and in a form set by the institution's risk assessment, with tests run or reviewed by independent third parties or independent staff. They do not name a penetration test or set an annual interval.
An InfoSight external test leaves you with:
It depends on how many live addresses and exposed services are in scope. Vendors that publish time estimates put an external network test at about 2 to 5 testing days, which is tester time rather than calendar time; the calendar length, the testing window and what reporting and re-testing include are agreed during scoping.
It depends on the rule that applies to you. PCI DSS sets internal and external tests at least every 12 months and after significant changes; the FTC Safeguards Rule calls for an annual penetration test unless you run effective continuous monitoring or other ongoing vulnerability-detection systems, and your risk assessment sets which systems it covers, with the rule's definition allowing testing from outside or inside; NYDFS Part 500 requires one at least annually for covered entities that are not exempt; HIPAA requires a periodic evaluation without naming a test. Major releases, acquisitions and cloud migrations are the kind of change a risk assessment weighs when deciding to test in between; outside those rules, your risk assessment and your rate of change set the cadence.
Testing stays inside a scope you approve up front, against systems you designate, with every action documented. Anything that must not be touched is written into the rules of engagement as off limits, and windows can run off-peak. Those safeguards, along with excluding disruptive techniques, reduce operational risk but do not eliminate it, so the engagement names a stop-work contact who can pause the test.
No. PCI DSS Requirement 11.4.3 calls for external penetration testing at least once every 12 months and after any significant infrastructure or application upgrade or change. A scan does not replace a test: NIST describes penetration testing as evaluators mimicking real-world attacks, often combining vulnerabilities to gain more access than any single one allows.
Yes, within each provider's rules. AWS lets customers test its listed services without prior approval and prohibits denial-of-service testing, real or simulated, outside its separate DDoS simulation policy. Microsoft no longer requires pre-approval for Azure but requires its penetration testing rules of engagement to be followed. Google Cloud does not require you to contact it, but your tests must affect only your own projects. For configuration inside those accounts, see InfoSight's cloud security assessment.
Yes. InfoSight uses U.S.-only penetration testers who hold certifications such as CISSP, OSCP and CEH, and does not outsource testing offshore, which supports data sovereignty, faster escalation and deeper manual exploitation.
Findings move into the Mitigator portal, each with a remediation status. After you patch, click Re-Test and InfoSight re-attacks those CVEs within 24 hours to validate the fix. PCI DSS also expects exploitable findings to be corrected and the test repeated to verify them.
Book a zero-cost 15-minute scoping call and instantly receive a preliminary scan report to share with stakeholders and executives.
No spam—one expert follow‑up, guaranteed.
Stay informed of the latest cyber trends.