Wednesday, August 9, 2006

Google AdWords and Click Fraud

The recent back-and-forth between Google and third-party firms estimating the extent of click-fraud in pay-per-click advertising is very interesting to me. As a blogger whose blog is hosted by blogspot, a Google subsidiary, and who derives income from Google ads on which some of my readers find interesting enough to click once in a while, it is in my interest for advertisers to feel confident that there aren't any shenanigans going on. (I also rotate the Google ads with ads from other sources, some of which are pay-per-click, some of which are pay-per-impression, and others of which are based on flat rates for periods of time.)

Google has come out swinging at its AdWords blog with some specific examples of flaws in the estimate put forward by Outsell that click fraud is approaching fifteen percent:

One clear indication that the consultants' results are flawed: they'’re not even getting the total number of clicks correct. We have seen some instances of reports showing 1.5 times the number of clicks in our logs --– for example, in one case 1,278 clicks were claimed as being "fraudulent" by the consultant while only 850 actually even appeared as clicks in Google's logs.

More evidence of the consultants'’ defective methodology is revealed when looking at conversion rates. We found clicks identified as "fraudulent" in reports often converted at nearly the same rate (and in some cases better) compared to other clicks. In one case, "fraudulent clicks" converted 5.1% of the time -- only a bit less than the advertiser's overall conversion rate of 5.8%.

So Google sez, you're showing us more fraudulent clicks than we can find clicks in this particular account, period. And those "fraudulent" clicks are "converting" (that is, arriving at a predetermined page on the advertiser's website, usually the order form) at very nearly the same rate as "authentic" clicks.

I have dabbled in advertising this blog through AdWords a time or two. A couple of times, I advertised on specific sites whose readers I felt might have an affinity for my rants. Other times, I just bid on key words that people might be searching for, or that might appear on other sites featuring Google ads. In my case, every click was a "conversion," since the point was to get the clicker to check out my blog. In checking out the activity of those who came to my site from clicking on the ads, I found that a good number of them made multiple page views. I have no idea how many of them ever came back, but I have to assume that some of them are now regular readers.

The one thing that I would like to know is, in the case of my ads that appeared on sites not of my choosing, what exactly were the sites? When I've clicked on links from sitemeter that are labeled as having brought visitors to me via Google ads, I've only been able to see a mockup of how the ad appeared, and the adjacent ads. I'm not given the context of the ad, that is, what website it appeared on. As an advertiser, I'd have more faith in the process if I knew where the ads were being published. Advertisers in every other medium know where their ads are appearing: on which radio stations, in which newspapers, or on which TV shows. As an AdWords customer, I usually have no idea.

I would think Google would want to provide this information to me, so I can try and secure advertising spots on sites that give me good traffic. I would do this by bidding a higher amount per click on sites that I find appealing. Google gets more money, the publisher gets more money, and I get more visitors.



Linking to the following blogs with open trackback posts: Third World County, Stop the ACLU, Diane's Stuff, Stuck on Stupid, Cigar Intelligence Agency