Most attacks do not work the first time.
That is expected.
What matters is not the initial failure, but what follows. Attackers observe what stopped them, adjust their approach, and try again, often quietly and patiently.
This learning loop is one of the most underestimated risks organizations face.
Why failure rarely discourages attackers
Defensive controls are visible in their effects.
When an attempt fails, attackers learn:
1. Which controls exist
2. Which alerts are triggered
3. Which actions are blocked versus ignored
Every failed attempt provides feedback.
That feedback improves the next attempt.
Adjusting the method, not the goal
Attackers usually keep the same objective.
They change:
1. The entry point
2. The timing
3. The person targeted
4. The communication channel
If email fails, they try messaging platforms.
If one employee resists, they approach another.
The destination stays the same.
Organizational behavior influences how infidels respond by
1. Paying particular attention to the speed of password resets.
2. The extent to which controls are relaxed after their discovery.
3. The order in which teams responded; and how long it takes for their vigilance levels to decline.
Most attackers return for additional attacks after controls have been relaxed rather than immediately after they have successfully gotten into the system.
Once an initial loud attack is made, infidels reduce their noise level and become more sophisticated. As infidels continue to succeed in their attempts, they will transition to being more focused, deliberate and methodical in their follow-ups and less easily recognized from legitimate actions.
When a successful attack ultimately occurs, it will likely appear to be a routine or normal business activity.
Real-world examples
Example 1: A Failed Phishing Campaign
An initial phishing compromise was reported and stopped.
What changed:
1. The next attempt was sent to a smaller group of recipients
2. Language was similar to the way that they usually communicate internally.
3. Timing coincided with a real project that was underway.
Result:
There was only one response.
Example 2: Blocked login attempts
Credential stuffing was detected and stopped.
What changed:
1. Attempts slowed dramatically
2. Logins occurred during business hours
3. Access came from expected regions
Result:
Detection thresholds were avoided.
Example 3: Denied system access
An attacker lost access to a system.
What changed:
1. Focus shifted to third-party access
2. A vendor account was targeted instead
Result:
Entry was regained without triggering the same controls.
Why this matters to leadership
Security is often measured by what was blocked.
Attackers measure something else:
What was learned
A failed attack today can inform a successful one months later. This creates delayed risk that is hard to attribute to a single event.
What reduces adaptive advantage
Organizations that limit attacker learning tend to:
1. Vary controls instead of relying on static thresholds
2. Treat minor incidents as signals, not noise
3. Review what a failed attempt revealed
4. Avoid predictable responses
5. Maintain controls after attention fades
The goal is not perfection.
It is reducing what failure teaches an attacker.
Questions leaders should ask
1. What does a failed attack reveal about us?
2. Which controls signal their presence clearly?
3. How long do we maintain heightened awareness?
4. Where do we assume attackers will give up?
These questions focus on resilience, not just prevention.
In conclusion, attackers become better at their trade through safe failure. Each failed attempt enhances their knowledge of an organization over time, transforming an organization's defenses into informative signposts. By reducing their risk, organizations will also eliminate most opportunities attackers have for learning. As attackers have less opportunity to learn from their failures, the opportunity for them to succeed will be greatly diminished as well.