In recent months, there has been considerable debate about the effectiveness of third-party click fraud auditing services.
A recent paper published by Google's Click Quality Team identified
several egregious flaws in the way that third-party click auditing
firms were identifying fraud. However, recognizing that
advertisers need an independent way to verify the validity of
pay-per-click charges, Google made several recommendations to
legitimately detect suspicious clicks:
- use the AdWords auto-tagging feature to match up which clicks came from Google as "unique" respondents;
- in particular, make sure that page reloads are not accidentally counted as clicks;
- when click fraud is suspected, an audit trail
of exact "click events" is much more helpful than sample-based analysis
of "estimated" click fraud to corroborate the truth.
In our opinion, these are good recommendations, but
they're just the start. ion offers services and tools to help
companies deploy highly effective landing pages and landing
"paths". Since many of the respondents we handle come from
pay-per-click vehicles such as Google, we have run into click fraud
many times: both malicious and unintentional. Through this
experience, we have developed several techniques for accurately
detecting suspicious traffic.
We've found that the best way to identify potential click fraud is to
analyze traffic as a "first-party" - the one actually serving the pages
to respondents - using techniques that would be difficult to implement as
a third-party.
If you're researching what strategies you might use to monitor click
fraud in your search engine marketing, we think you will find some of
our ideas useful. Click here to learn more about "Catching Click Fraud with ion's Post-Click Marketing System."
- Scott Brinker