Skip to content
English
Contact us
  • There are no suggestions because the search field is empty.

Audit Rules & Traffic Classification

This guide explains how Vaudit identifies and classifies different types of ad traffic. It outlines the rules behind our audit decisions and helps you understand which events are flagged, why they're flagged, and what standards they align with.

Overview of Classification

Vaudit flags traffic using a hierarchy of rules, based on the strength of evidence for suspicious or non-compliant behavior:

  • Invalid Traffic: Strong indicators of invalid or non-compliant traffic. These are often used to support refund claims and prioritised for real-time charge removals. 
  • Suspicious Traffic: Weaker indicators of low-quality traffic. These may not qualify for refunds but can guide some charge removals (e.g. user exclusions or signal adjustments).

Traffic Rules and Definitions in the correct order of hierarchy:  

Invalid

1. Non Compliant Traffic Sources:
Non-Human (Invalid) 

Definition:
Non-compliant traffic sources include automated bots, fraudulent actors, or suspicious activity that mimic real users but violate platform policies and deliver no legitimate engagement or value. These sources artificially inflate your clicks, impressions, and spend.

Why it’s flagged:
We flag these sources based on a combination of signals:

  • Detection of headless browsers or bot user agents.

  • Matching IPs or user agents against the IAB’s known bot list.

  • High fraud risk scores (≥75%) from Pixelate API.

  • Behavioral patterns such as 3 or more different click IDs (GCLID/FBCLID) from the same IP within 60 seconds.

Standard:
Supported by IAB Bot Detection Guidelines and the TAG Fraud Framework, which recognize abnormal browsing patterns and known bot signatures as reliable indicators of invalid, non-human traffic.

 2. Non-Compliant Traffic Sources:
Zero-Second Clicks (Invalid)

Definition: Clicks that result in no pageview.

  • Why it’s flagged: A user clicked your ad but never loaded the content of the page.
  • Possible causes: Broken tracking, bounce blockers, or click fraud.
  • Standard: Aligns with MRC viewability principles and Google policies on invalid clicks.

3. Geo Violations (Invalid)

Definition: Clicks that came from geographic regions you've actively excluded in your campaign settings.

  • Why it’s flagged: These clicks directly violate your targeting preferences.
  • Standard: Supported by IAB, MRC, TAG, and Google policies as invalid traffic.

4. Non-Compliant Traffic Sources:
High Frequency Clicks (Invalid)

  • Why it’s flagged: Highly suspicious behavior indicating non-human activity.
  • Standard: While the industry standard (MRC/IAB) flags suspicious activity if a single IP loads 3 or more pages in 30 seconds, our system uses a much stricter threshold: 3+ page loads in under 10 seconds. This means that if traffic is flagged by our system, it is exhibiting extremely rapid, non-human behavior—far beyond what is considered suspicious by industry norms.

5. Device Violations

Definition:
Device Violations refer to discrepancies and suspicious patterns related to device types, platforms, and user agents that suggest automated or spoofed activity, including device fingerprint diversity, device/browser inconsistencies, and rapid switching between devices from a single IP.

Why it’s flagged:
We flag device violations when:

  • An IP exhibits unusually high diversity of user agents in a short time (e.g., 5+ user agents in under 30 minutes).

  • A device claim (e.g., “iPhone”) does not match its user agent string (platform spoofing).

  • An IP rapidly rotates between significantly different device types (e.g., mobile, desktop, tablet) within 10 minutes.

  • User agents show bot-like signatures (e.g., “HeadlessChrome”, “PhantomJS”, "Selenium") or abnormal lengths.

These behaviors indicate attempts to disguise fraudulent activity by impersonating multiple devices.

Standard:
Supported by MRC IVT 2.0 Standards for Device/Environment mismatch and consistent with best practices for identifying emulated, spoofed, or automated device behavior in advertising environments.

6. Session Violations (Invalid)

Definition: Clicks where users land on your page but don’t interact at all.

  • Why it’s flagged: No clicks or interactions signal low engagement or accidental/invalid visits.
  • Standard: MRC and IAB recognize bounce patterns as potential signals of invalid traffic.

Suspicious

7. Session Violations (Suspicous)

Definition: Same as above, but evaluated against an acceptable bounce rate.

  • Why it’s flagged: When your actual bounce rate exceeds a threshold, we calculate how much you may have overpaid.

Campaign Type

Max Bounce Rate (Invalid)

Max Bounce Rate (Suspicious)

Search/DemandGen

50%

30%

PMax/Display/Shopping/Video/Smart

70%

50%

8. Geo Violations (Suspicious)

Definition: Clicks from areas that weren’t specifically excluded but also weren’t targeted.

  • Why it’s flagged: These locations fall outside your intended business region and may underperform.
  • Standard: Not invalid by definition, but often considered lower quality.


9. Non-Compliant Traffic Sources: 
High Frequency Clicks (Suspicious)

Definition: Same IP generates 5+ page loads in under 60 seconds.

  • Why it’s flagged: Indicates potential abuse or repeated clicks from a single source.

10. Low Activity Hours Clicks (Suspicious)

Definition:
Google Ads clicks that occur between 2:00 AM and 6:00 AM local time (based on your website’s time zone).

  • Why it’s flagged:
    This time window typically sees low user intent, higher risk of accidental clicks, or bot-driven activity. These clicks are unlikely to result in meaningful engagement or conversions, and may represent wasted spend.
  • Standard:
    While no specific MRC or IAB standards govern hourly targeting, this approach aligns with best practices in invalid traffic detection and is widely used by performance advertisers to reduce off-peak inefficiencies.

How We Calculate Impact

For each flagged event, we calculate the estimated overcharge or refund amount using:

  • The number of affected sessions or clicks
  • The average cost per click (CPC) for the campaign

These amounts are rolled up in your audit dashboard, giving you clear visibility into what’s happening, and where optimizations or refund requests might be warranted.


Why This Matters

These audit rules ensure that:

  • You only pay for valid, high-quality traffic
  • You have clear documentation to support refund discussions with platforms
  • You get actionable insights to reduce waste and improve campaign ROI

If you have any questions about how specific rules apply to your account, feel free to reach out to your Vaudit rep.