Which AI Score bands are suspicious clicks?

ClickPatrol labels clicks as suspicious when they fall into the somewhat medium, medium, or high AI Score bands. Safe and low clicks are treated as broadly consistent with normal traffic. The ladder lets teams prioritize review on upper bands first while keeping transparency about which interactions need scrutiny versus automated handling.

Are all suspicious clicks automatically blocked?

ClickPatrol deliberately does not block every suspicious click. Blocking happens only when the system is confident the interaction is fraudulent or non-human in a way that will not cost real conversions. Borderline cases may be monitored, scored, or allowed depending on account settings and historical conversion behavior to reduce false positives.

Can you change how aggressively clicks are blocked?

Advertisers can adjust blocking aggressiveness in the ClickPatrol dashboard. Stricter blocking may save more spend but raises the risk of excluding legitimate users on risky networks or devices. Default settings aim to avoid blocking clicks that still carry conversion value, balancing fraud reduction against potential lost legitimate sessions.

How does ClickPatrol score a click as suspicious?

When a click arrives, ClickPatrol enriches it with Google and other data sources, then scores hundreds of factors including device, network, timing, and consistency with declared geography. Results roll into AI Score bands from safe through high. Suspicious classification supports campaign-level visibility into click fraud pressure without treating every flagged click as billable fraud.

What are suspicious clicks?

Abisola | Sep 30, 2024

In ClickPatrol, suspicious clicks are interactions we classify as potentially non-human, low quality, or otherwise risky after we analyze each click against our models and third-party intelligence. They are not yet a verdict on billing; they are a strong signal that the click deserves scrutiny or automated handling.

How we identify suspicious clicks

When a click arrives, we enrich it with data from Google and other sources, then score it. Our system evaluates hundreds of factors (device, network, timing, consistency with declared geography, and more) to estimate how human-like the session is. The result rolls up into AI Score bands from safe through high.

We label clicks as suspicious when they fall into the somewhat medium, medium, or high bands. Safe and low clicks are treated as broadly consistent with normal traffic. That ladder lets you skim dashboards quickly: focus on the upper bands first.

Suspicious does not always mean blocked

We deliberately do not block every suspicious click. We only block when we are confident the interaction is fraudulent or non-human in a way that will not cost you real conversions. Borderline cases may be monitored, scored, or allowed depending on account settings and historical conversion behavior.

You can adjust aggressiveness in the dashboard. Stricter blocking can save more spend but raises the risk of excluding legitimate users who share risky networks or devices. Default settings aim to avoid blocking clicks that carry conversion value. If you want the tradeoffs in plain language, read could you lose conversions with ClickPatrol.

Why distinguishing suspicious clicks matters

Advertisers need both transparency and safety. Raw click counts hide mixed traffic; separating suspicious activity shows where click fraud pressure sits by campaign, geography, or device. It also supports conversations with stakeholders about why certain IPs or patterns were excluded.

For deeper behavioral context, see what is suspicious behavior and how we detect fraud.

Frequently Asked Questions

Which AI Score bands are suspicious clicks?
ClickPatrol labels clicks as suspicious when they fall into the somewhat medium, medium, or high AI Score bands. Safe and low clicks are treated as broadly consistent with normal traffic. The ladder lets teams prioritize review on upper bands first while keeping transparency about which interactions need scrutiny versus automated handling.
Are all suspicious clicks automatically blocked?
ClickPatrol deliberately does not block every suspicious click. Blocking happens only when the system is confident the interaction is fraudulent or non-human in a way that will not cost real conversions. Borderline cases may be monitored, scored, or allowed depending on account settings and historical conversion behavior to reduce false positives.
Can you change how aggressively clicks are blocked?
Advertisers can adjust blocking aggressiveness in the ClickPatrol dashboard. Stricter blocking may save more spend but raises the risk of excluding legitimate users on risky networks or devices. Default settings aim to avoid blocking clicks that still carry conversion value, balancing fraud reduction against potential lost legitimate sessions.
How does ClickPatrol score a click as suspicious?
When a click arrives, ClickPatrol enriches it with Google and other data sources, then scores hundreds of factors including device, network, timing, and consistency with declared geography. Results roll into AI Score bands from safe through high. Suspicious classification supports campaign-level visibility into click fraud pressure without treating every flagged click as billable fraud.

Abisola

Abisola handles content and support at ClickPatrol. She helps customers get more value from cleaner traffic data and writes practical resources about ad fraud, fake traffic, and smarter PPC decisions.