Skip to main content

FraudSensor Domain Mismatch IVT Classifications FAQ

These are common questions related to FraudSensor domain mismatches. For more information on how HUMAN classifies domain mismatches, see Domain Mismatch IVT Classifications.

What should I do if domain mismatches occur?​

If you see domain mismatches on your traffic that have been identified as either invalid or valid traffic, then we recommend you first check your HUMAN integration to make sure it is not an integration issue.

After you verify it is not an integration issue, then any domain mismatches that have been identified as invalid traffic should be treated the same way you would any other invalid traffic. You should still investigate domain mismatches classified as valid traffic to try to identify and fix the issue.

What should I check in my FraudSensor integration?​

You should check that the Domain Intended parameter is properly populated and includes both root and subdomains. Domain Intended is the domain where the ad was meant to be delivered or where the event or impression will occur. This value should include the full top-level domain from the URL in the site.page ORTB field, including any subdomain parts, but should exclude the protocol and path.

The Domain Intended parameter is one of the required parameters that must be provided during the FraudSensor Tag Integration, and the FraudSensor Mismatch technique relies on this parameter.

It's necessary to ensure that the parameter is populated properly and includes both root and subdomains.

When will mismatches be flagged?​

Domain mismatches will be flagged differently across different HUMAN features under different conditions. The flags and their conditions are:

  • FraudSensor SIVT tag: Applied when FraudSensor identifies a material mismatch.
  • FraudSensor Domain Mismatch: A metric available in the FraudSensor SnowFlake Data Share: domains.mismatch and FraudSensor Explore: Domain Mismatch. It is set to true per FraudSensor transaction whenever it identifies a mismatch between the intended domain and detected domain. This includes root and subdomains as well as valid or invalid traffic.
  • Domain Mismatch Insights: A metric that identifies the most common mismatches between the MediaGuard declared and FraudSensor detected domains. This includes root and subdomains. This metric is available on the Insights (Beta) and Explore pages in the HUMAN Dashboard.

Are Domain Mismatch IVT Flagging Indicators available in reporting?​

No, indicators described in Domain Mismatch IVT Classifications are not surfaced in reporting today. However, domain mismatches classified as IVT can be reviewed using the Domain Mismatch flag and the IVT subcategory Domain Spoofing in the HUMAN Dashboard.

Will this impact MediaGuard predictions?​

No, this detection technique has been specifically excluded from propagating within the MediaGuard model. As we progress with our detection in this area, we will look to expand our coverage to MediaGuard.