RKG Logo 434-978-4300

Everyone’s talking about clickfraud recently. John Battelle. Robert Cringley. Andrew Goodman. Heck, even Mark Cuban.

But not enough folks emphasize the key distinction: search click fraud vs. content click fraud. Follow the money. Content click fraud is the key issue.

Though not limited to Google, consider AdWords and AdSense. With 43% share, Google’s a reasonable proxy for the industry.

  AdWords Fraud AdSense Fraud
Harms Advertiser Advertiser
Benefits Google
Advertiser’s competition (a bit)
* * AdSense publisher * *
Google
Advertiser’s competiton (a bit)
Scaleable Inventory No Yes
Scaleable Fraud No Yes
“Death By 1000 cuts”
API

Consider AdWords fraud, where a person or robot clicks on a search result ad on Google or a syndication partner for the sole purpose of incurring the click charge.

Clearly, the advertiser is harmed. Who benefits? Well, Google makes a few cents. And the advertiser’s competitors gain slight benefit from the advertiser’s loss of a bit of their marketing budget. But both Google and any competitor of size have too much to lose and too little to gain by such mischief. There’s too much risk to build a clickbot or hire a firm to click fraudulently.
Besides, I believe that Google and major retailers act ethically.

Detecting bogus AdWords clicks is hard but possible, and I think Google does a good job detecting and purging them. And the AdWords click universe scales (relatively) slowly, driven by Google seizing market share, the increasing availability of broadband, and the growing adoption of search. The growth is limited to real people clicking on real search results. Clicks aren’t generated out of thin air. As a result, AdWords fraud scales slowly, both in relative and absolute terms.

In contrast, consider AdSense fraud, where a person or robot clicks on a context ad on an AdSense publisher site for the sole purpose of incurring the click charge.

As before, the advertiser is harmed. But who benefits? Again, Google makes a few cents. But more importantly, the publisher makes a few cents. And unlike AdWords fraud, AdSense fraud scales extremely well — the publisher can use scripts to create numerous sites easily (even more so now with the AdSense API). Across many many sites, a few cents here and there can add up to meaningful easy money. As the fraud can be spread across many AdSense accounts, it is much harder to detect, even with legions of PhDs. The AdSense click universe can grow quickly, driven by manufactured clicks on fake sites.

In general, I believe AdWords fraud rates are in the low single digits, and
AdSense fraud reaches as high as thirty percent.

From the retailer perspective, if the ads drive enough sales to meet target ROI, then the retailer should run those ads, search and/or content. The click fraud loss is just a cost of doing business. (Of course, an advertiser would be crazy to run search and content without distinguishing tracking.)

But one thing is sure: when analyzing click fraud, the search vs. content distinction is crucial.

If you like this post, consider subscribing to our RSS feed. You can also have new posts sent to you via email.


Possibly Similar Posts

Trackback

http://www.rimmkaufman.com/rkgblog/2006/06/07/adwords-vs-adsense-click-fraud-why-doesnt-anyone-get-it/trackback/

Blogs Citing This Post

  1. Pingback: Itemized Click Reports From Search Engines To Advertisers Would Increase Transparency and Combat Click Fraud on December 20, 2007
  2. Pingback: AdSense: Crossing To The Dark Side on August 11, 2008

Comments

  1. Mattg, January 13, 2007:

    Good article have lot of info.

  2. SirCommy, December 12, 2007:

    Well said. I am running adwords but not adsense. I have been running adsense long time ago though. But when it comes to adwords, fraud clicks are on a much larger scale for the content ones. I chose my ad to display only on search and not on content or network partner sites. Those are the places where most fraud clicks are being done. For example, when I have content and partner sites turned ON, I was having over 10 clicks per 2 hours and no interested visitor. That’s because most earn from adsense when content ads are being clicked. So after I turned it off and left it for search only (sponsored ads on searches), I had few clicks, but real visitors and subscribers. The difference is huge. I am interested on the type of visitors that search specific terms and get my site, not intentionally displayed on I don’t know what type of sites.
    Good article.

  3. Adsense, September 30, 2008:

    Yes your comments are spot on. I hate the possibility of fraud and hope I never get accused by accident. If someone hates you enough they can cause you some damage by clicking on your ads, even in Adsense or Adwords.
    Regards,
    Andy

Your Comment

Tags

RKG Tags:

Technorati Tags:

Email Updates

Categories

Recent Comments

  • George Michie: Happy New Year, John. Market Motive is a good place to start. Keep a close eye on other good blogs: ours, SearchEngineLand, ClickZ....
  • john: Hi George and Happy New Year. I work for large Company who created a new SBU of New Media Specialists, basically resellers of Google And...
  • George Michie: Shelley, I hear you! We had a candidate who listed HTML as a proficiency on her resume. When pressed for specifics she said:...
  • Erin: Actually, I am always wondering this about myself. I surprise myself by what I know, but I don’t feel super confident about my...
  • Shelley Ellis: I like the sports analogy. My husband was a baseball coach for years and I still get tickled thinking about the time a mom told him...
  • AJ @ Web Domains, UK: I think the problem is spamming scripts are being written that target prominent blogging/comments systems such as Movable...
  • Debra Askanase: Hi Alan, Thanks for the quick overview. Just want to point out that the Wufoo outgoing link actually connects to Survey Monkey....
  • George: Vijay, that’s the best idea I’ve heard in months! Rick, your point is well-taken. It’s really noisy, and for many retail...
  • Alan: You can use this toy model in any channel where you can sales definitively tie sales back to an advertisement or promotion using a tracking...
  • Patrick: Nice model. What I’m wondering however is what type of net sales you should include. I believe these should be the sales resulting...
  • Vijay R: With mobile phones being ubiquitous, why not use them as a channel to measure effect of online on offline sales? For instance, offer a...
  • George: Not all of our clients do think this way. Many do, but others are more concerned about share of voice, and the trappings of a good program....
  • Jim Novo: George, I have to ask the same question I asked Alan on his “melons” post - how do we get people to care about profits? Every...
  • Rick Galan: Even doing specific geo-targeted tests really don’t get us all of the way there though. There is so much fluctuation in consumer...
  • Andrew: $800 billion is perhaps “only” 5% of GDP, but it exceeds all of the federal government’s other (individual) outlays....

Blog Stats

  • Posts: 802
  • Words: 356,639
  • Comments: 1,584

Administration