Press TechRound interviews Secure.com CEO on the future of AI security
Read

Why Annual Pentesting Is Broken

84% of organizations were breached in 2024, yet most only pentest once a year. Here is why annual pentesting fails and what to do instead.

Key Takeaways

  • 84% of organizations were breached in 2024. Only 26% ran more than one pentest that year.
  • Attackers exploit vulnerabilities in 5 days on average — your next annual test is months away.
  • Cloud environments change daily. Developers push new services, APIs go live, and buckets get misconfigured. Your last pentest knows nothing about it.
  • The security gap year is the 364 days between annual tests. Attackers treat it as open season.
  • One breach costs a startup an average of $3.31 million. For many, that is not a setback — it ends the business.
  • Continuous and monthly testing closes the gap without requiring a bigger team.

Introduction

84% of organizations were breached in 2024. Most ran exactly one penetration test that year. That math does not work in anyone’s favor.

Annual pentesting was built for a slower era. The threat landscape today moves daily. Here is what is actually going wrong — and what a better approach looks like.

The Security Gap Year

365 days in a year. Your annual pentest covers 1.

Each dot below is one day. Attackers work every single one.

Pentest day
Unprotected
1
Day your infrastructure is actively tested each year
364
Days attackers have to find what you missed

Why Do Companies Get Breached Between Annual Pentests

Every company that finishes an annual pentest walks away with a clean report and a sense of relief. That report was accurate on one day. The other 364 days are a different story.

Breaches rarely happen during a test. They happen in the window between tests — when a new misconfiguration goes unnoticed, when a forgotten subdomain gets picked up by an automated scanner, when a developer spins up a service nobody added to the asset inventory.

Attackers do not wait for your next engagement. They scan for new exposures the moment they appear.

How Fast Do AI-Enabled Attackers Exploit External Vulnerabilities

AI-Enabled Attackers

Attackers used to take a month to exploit a vulnerability. Now it takes 5 days.

Your annual test is already outdated before the report lands in your inbox.

2022 average
32
days
2025 average
5
days
84% faster
Speed
28%
of vulnerabilities exploited within 24 hrs of disclosure
AI accuracy
87%
success rate when AI is given only a CVE description
Cost to attacker
$1
per AI-assisted exploit attempt against your infrastructure

The average time to exploit a disclosed vulnerability dropped from 32 days in 2022 to just 5 days in 2025. That is not a slow, methodical process. That is a near-instant attack window.

Mandiant's research found that 28% of vulnerabilities are now exploited within 24 hours of disclosure. Some CVEs are being attacked before a patch even exists — a concept researchers now call "negative time-to-exploit."

AI tools are a major driver of this. GPT-4 can exploit documented vulnerabilities with an 87% success rate when given only a CVE description. At roughly $1 per exploit attempt, attackers run thousands of tests against exposed infrastructure at near-zero cost. The economics now heavily favor the attacker.

Why Annual Pentests Miss Critical External Exposure

Over 40% of CISOs report that their pentest results are already invalid by the time the final report lands in their inbox. That is not a technology problem. It is a timing problem.

New services go live after the test. API endpoints get deployed without a security review. Old certificates expire and get replaced with misconfigured ones. Dev environments go online for a sprint and never come back down.

The pentest you ran in January knows nothing about the infrastructure you built in July. And that is exactly where attackers look first.

Why Most Companies Miss the Security Gap Year

Nobody officially names it, but the security gap year is real. It's the stretch of time between your last pentest and your next one. For most companies, that stretch is 12 months.

For attackers, that gap's an open invitation.

Over 21,000 new vulnerabilities were disclosed in 2023 alone — roughly 57 per day, according to CVE database records. Your annual test covered zero of them the moment it was complete. The attack surface your team signed off on is, by definition, already stale.

How Attackers Exploit the Security Gap Year

Attackers perform their own reconnaissance constantly. Tools like Shodan and Censys map your external exposure in real time. They do not need a scheduled engagement. They build a live picture of what you expose to the internet and watch for it to change.

New CVEs get weaponized fast. Automated exploit frameworks test known vulnerabilities across thousands of targets simultaneously. If your infrastructure is reachable and unpatched, it shows up in those results — no matter when your last test was.

The gap year is not a theoretical risk. It is the documented reality behind most modern breaches.

What the Gap Year Means for Startup Security

For startups and SMBs, this hits harder. 43% of all cyberattacks target small businesses, according to recent cybersecurity research. A single breach costs an average of $3.31 million for companies with fewer than 500 employees, according to IBM's 2025 Cost of a Data Breach Report.

60% of small businesses that suffer a cyberattack close within six months, according to industry research. One gap-year breach — one misconfiguration found in month nine of your testing window — can end the business entirely. Annual testing doesn't protect against that.

Why Does Attack Surface Change Faster Than Annual Assessments

Your cloud infrastructure does not sit still. A developer can spin up an EC2 instance, create an S3 bucket, deploy an API, or add a subdomain in minutes — without a security review, without a ticket, and without updating the asset inventory from your last pentest.

The moment workloads move to the cloud, the scope document from your last annual test is already incomplete.

How Attackers Discover Exposed Cloud Resources

Attackers use automated tools to find exposed buckets, forgotten subdomains, and misconfigured services. S3Scanner, Shodan, and similar tools scan the public internet continuously and flag anything reachable without authentication.

Cloud-focused attacks jumped 37% in 2025, according to recent threat intelligence reports. Research confirmed that nearly half of all AWS S3 buckets analyzed in security assessments showed potential misconfiguration risks — most caused by default settings, permission changes, and test environments that never got cleaned up.

One open bucket found during a scan can expose customer records, internal credentials, and source code. That exposure can exist for months before anyone inside the company notices. Automated external scanners used by attackers notice it in minutes.

Attack Surface Drift

Your cloud changes every week. Your annual pentest does not know about any of it.

These are the things that commonly shift between tests — each one a potential attacker entry point.

🪣
High risk
S3 buckets created and forgotten

Developers spin up storage for a sprint and never decommission it. Default permissions often leave data publicly accessible.

🔌
High risk
APIs deployed without security review

New endpoints go live between release cycles with no auth check, no pentest scope update, and no inventory entry.

🌐
Medium risk
Subdomains pointing to dead resources

DNS records outlive the services they point to. Stale subdomains are a common subdomain takeover target.

🔑
High risk
IAM roles quietly expanding

Permission creep happens gradually. A role added for a one-off task stays active and over-privileged indefinitely.

🖥️
Medium risk
Dev environments left running

Test and staging instances stay online past their purpose with weaker configs, default passwords, and no monitoring.

📦
High risk
New CVEs released daily

Over 21,000 vulnerabilities were disclosed in 2023 alone. Your annual test covered zero of them after it was complete.

What Is the Hacker's View of Your Attack Surface

Attackers look at your environment from the outside in. They don't care about your internal architecture, your ticketing system, or how your teams are organized. They care about anything reachable from the internet.

That includes forgotten subdomains, exposed admin panels, API endpoints without authentication, misconfigured cloud storage, and public repos with hardcoded credentials. Your annual pentest may have reviewed a clean inventory at one point in time. But configurations drift. New things get built. And nobody is watching the external surface in between.

Why Is Monthly Pentest Cadence Better Than Annual

Testing Cadence

Annual testing is a compliance checkbox. Monthly testing is an actual security practice.

Here is what that difference looks like in practice.

📅
Annual Pentest
Once a year
364 days of unmonitored exposure
Report is stale before it lands in your inbox
New cloud assets and configs not covered
Primarily a compliance checkbox, not a security tool
Misconfigurations go undetected for months
🔁
Continuous Testing
Monthly cadence
Max 30-day gap between assessments
Findings reflect current infrastructure state
New deployments and cloud drift caught fast
Continuous baseline gives teams a real security posture
Misconfigurations caught before attackers find them

Annual testing is a compliance exercise. Monthly testing is an actual security practice.

When you test monthly, you catch misconfigurations before attackers do. You verify that last quarter's deployment did not open a new hole. You give your team a current baseline instead of a 12-month-old snapshot.

The shift is not complicated. The payoff is significant.

Why Penetration Testing Cadence Matters Now

The threat environment has changed permanently. AI-assisted attacks mean vulnerabilities get weaponized in days. Cloud adoption means your attack surface changes faster than most IT teams track. The window between exposure and exploitation is now measured in hours, not months.

Quarterly testing isn't enough. Research found that attackers can achieve critical compromise in as little as 60 seconds once they have a foothold in cloud environments. The question is never whether someone will look. The question is how long they have to find the door.

Monthly testing removes most of that window.

How Do Lean Security Teams Manage External Exposure

Most startups do not have a dedicated 10-person security team. They have one or two people who own security alongside three other responsibilities.

That is precisely why continuous and automated external testing matters more for small teams, not less. Modern infrastructure security tools run recurring scans, flag new exposures automatically, and surface what changed since the last scan — without requiring a manual review of every configuration across every service.

The goal is not to test everything at once. It is to make sure that nothing new stays exposed for long.

Stop Relying on Snapshots. Your Infrastructure Changes Daily.

Infrastructure Security Teammate
Your attack surface does not take days off. Neither should your security.

Annual testing gives you a snapshot from one day. The Infrastructure Security Teammate watches your infrastructure every day — the way attackers do.

Annual pentest: 364-day blind spot
Teammate: caught in minutes
<5 min
to detect a new misconfiguration
70 %+
reduction in manual triage workload
3  clouds
AWS · Azure · GCP
🗺️
Attack surface mapping

Every exposed asset — domains, cloud services, forgotten subdomains — mapped and updated in real time.

🔄
Continuous drift detection

When a bucket goes public or a port opens without authorization, your team knows in minutes.

Automated remediation

Findings are correlated and prioritized across cloud, endpoint, and identity layers — automatically.

👥
Built for lean teams

One or two people can maintain enterprise-level visibility. No extra headcount needed.

Stop finding out months after the fact.
See exactly what an attacker sees — before they act on it.
Explore the Infrastructure Security Teammate →
No agents required · Works with your existing stack · First value in 30 minutes

FAQs

What is the cost of undetected vulnerabilities for startups?
A single breach costs an SMB with fewer than 500 employees an average of $3.31 million, according to IBM's 2025 Cost of a Data Breach Report. That figure covers downtime, regulatory fines, legal costs, and reputational damage. For startups with limited cash, it is often unrecoverable - 60% of small businesses that get breached shut down within six months.
How do you set up monthly penetration testing scans?
Start with an external attack surface scan that covers your domains, subdomains, IP ranges, and cloud assets. Set a recurring schedule using automated tools that replicate attacker behavior from the outside in. Review findings monthly and prioritize anything with public exposure or access to sensitive data. Tools like Secure.com's Infrastructure Security (Cloud Security) Teammate are designed to automate this for teams without a dedicated security headcount.
How lean security teams manage external exposure?
Lean teams do best when they automate visibility instead of trying to manually review every change. Continuous external scanning flags new assets, misconfigurations, and configuration drift without requiring a full-time analyst on standby. Pair this with a clear remediation process and risk-based prioritization, and a small team can stay ahead of the most common external threats.
Why does attack surface change faster than annual assessments?
Cloud environments are dynamic by design. Developers provision new services without a formal security review. APIs go live, test buckets get created, and DNS records point to resources that no longer exist. Each one of those changes is a potential new attack vector. Annual assessments capture a single day. Your attack surface changes every week.

Conclusion

Annual pentesting made sense in a slower era. Today, it is a compliance checkbox that leaves most of the year completely untested.

Attackers exploit vulnerabilities in days. Cloud misconfigurations appear between sprints. AI tools run reconnaissance without stopping. A 12-month testing cycle does not match that reality — not for enterprise teams, and especially not for startups.

The fix is not just running more tests. It is having continuous visibility into your external attack surface so that new exposure gets caught before it becomes a breach.

Secure.com's Infrastructure Security Teammate brings that visibility to teams of any size. It continuously monitors your cloud attack surface, detects misconfigurations and drift in real time, and enables your team to remediate exposures before attackers exploit them — so the gap year becomes a thing of the past.