As part of the Lane Gifts settlement Google agreed to an independent review of their click fraud prevention program. Here is a link to the 47-page report, written by Dr. Alexander Tuzhilin, Professor of Information Systems at NYU. The report was filed with the court in Texarkana, Arkansas.
My take on all of this is that search engines need to do a reasonable job of filtering out invalid clicks to reduce it down to “noise level”, and suggest that advertisers who want deeper analysis of click stream data employ third party tools. Search engines could make it easier to use these tools by providing data feeds to advertisers that could be correlated to their own click stream analysis.
It is a competitive advantage to do a better job filtering out invalid clicks, just like it is a competitive advantage to provide better search results. However, in both cases, quality is highly subjective. Search engines need to take reasonable measures to filter out invalid clicks. It is understood that there is no way to filter out 100% of invalid clicks. Indeed, there is no generally accepted definition of what an invalid click is because it could be different from one advertiser to another, and even different for the same advertiser on different keywords, campaigns, or weeks of the year.
Click fraudsters are constantly adjusting their approach, just like spammers, to avoid detection and “fly under the radar”. For this reason search engines can’t say much about how they filter out invalid clicks and what attributes they look for.
Third party auditing has issues as well. There is no universal truth. Pick five different independent auditors and you could get five different opinions of click fraud in any given case. If the search engine pays the auditor there could be suspicion of a conflict of interest or influencing the results.
In the end CPC advertising, like all advertising, boils down to ROI. If the advertiser is happy with the ROI they will continue to advertise. Each advertiser has different thresholds for ROI happiness. Invalid clicks are just one small factor in the ROI equation.
From the Google blog,
Here are excerpts of some of the positive things Dr. Tuzhilin has to say about Google and invalid clicks:
“During this project, I visited Google campus three times and interviewed over a dozen of the Click Quality team members from the Spam Operations and the Engineering groups, as well as the Product Manager of the Trust and Safety Group. I found the members of both groups to be well-qualified and highly competent to perform their jobs. Most of them have relevant prior backgrounds and strong credentials.” (p.4)
“The current set of Google filters is fairly stable and only requires periodic 'tuning' and ‘maintenance’ rather than a radical re-engineering, even when major fraudulent attacks are launched against the Google Network.” (p.25)
“These inspection systems have been developed by Google over an extensive period of time and are constantly improved to extend their functionality and make them better for the investigators to do their inspections more effectively. I have personally observed several such inspections and can attest to how successfully they have been conducted by Google’s investigators. This success can be attributed to (a) the quality of the inspection tools, (b) the extensive experience and high levels of professionalism of the Click Quality inspectors, and (c) the existence of certain investigation processes, guidelines and procedures assisting the investigators in the inspection process.” (p. 40)
“Google has built the following four 'lines of defense' for detecting invalid clicks: pre-filtering, online filtering, automated offline detection and manual offline detection, in that order. Google deploys different detection methods in each of these stages: the rule-based and anomaly-based approaches in the pre-filtering and the filtering stages, the combination of all the three approaches in the automated offline detection stage, and the anomaly-based approach in the offline manual inspection stage. This deployment of different methods in different stages gives Google an opportunity to detect invalid clicks using alternative techniques and thus increases their chances of detecting more invalid clicks in one of these stages, preferably proactively in the early stages.” (p. 47)
Google filed a “response to objections” report that details why they think the settlement is fair, and why they think the Yahoo settlement is not. Interesting reading.
Subscribe - To get an automatic feed of all future posts subscribe here, or to receive them via email go here and enter your email address in the box in the right column.
Recent Comments