Slashdot Mirror
Webhost Sues Google
TheOcho writes "Webhost company AIT has decided to file a class action lawsuit against the internet giant Google. According to the article the dispute is over click fraud. AIT claims they have lost around $500,000 due to fraudulent clicks. They claim that Google is hitting their website from 'the same IP addresses'."
AIT stores
AIT launched its first storefront Thursday in Chicago.
The Fayetteville Web hosting company plans to open one or two stores each month in 2006 as part of a $5 million campaign to expand the company.
Convenient both are occuring at the same time so it can be mentioned in the same article. Looks like a news story then turns into a press release.
"My question to them is simple," Briggs said. "Don't you think you have a right to see which IP addresses you were charged for?"
Well they do have a point.
Google has this data, why not make it available?
If i were an advertiser I would want to be able to to verify that the bills Google sends me are indeed correct. Right now it seems that advertisers have no way of doing that?
But I can see why Google is reluctant, providing this data incurs more costs, and I can imagine that there will be a lot of advertisers who are going to argue with them about their bills.
Not surprisingly, the article is light on technical details. I don't believe that a corporation such as Google would seemingly overlook a simple address filter containing IP ranges used by known legitimate crawling agents.
Maybe spam agents were indexing the AIT Web sites in an effort to aggregate data like published e-mail addresses. The article just doesn't tell us much. If that were the case, however, Google wouldn't have many options. They could add end-user validation to each advertisement (i.e., "Repeat the alphanumeric string so that we know you aren't a robot!"), which would obviously inconvenience the user and ultimately decrease traffic, or they could create ban filters. I would suppose that the latter option might also garner various legal accusations.
It sounds as though AIT could have incurred a legitimate loss, but are pursuing a large corporation whose employees aren't exactly known by most people for their negligent behavior. If my suspicions are true, however, how could Google engineers manage to prevent "click fraud" while balancing the usability of their service? Nobody wants to spend thirty seconds validating themselves as a human to an advertisement. Maybe AIT would have better luck pursuing the (hypothetical) spammers.
Do you like German cars?
They can see IP addresses for clicks in their server referrer urls and thus they know that many are frauds (see slashdot et al passim for more info on fraudulent clicks).
Their complaint is effectively that google doesn't provide them with this info and so they have been asked to pay for X clicks when they would like to pay for Y distinct clicks.
They really have no case. Imagine a guy being paid to hand out leaflets in the street... suppose some other guy keeps walking back and forth taking a leaflet - is that the fault of the leaflet guy?
Justin.
You're only jealous cos the little penguins are talking to me.
With google, you pay when someone clicks on your link rather than how often the ad appears. You name your price for how much you'll pay for each click and that governs where and how often your ad appears. Obviously the more you'll pay for your keywords, the more Google will show it and the more impressions it will make. A part of that price you pay goes to the site that hosts it. Therefore I can reward a host site simply by clicking on their ads. I suppose in a way this classifies as "click fraud" since I rarely have the intention of buying whatever is being sold.
Another bonus (or detriment depending on your POV) of pay-per-click is that you can "punish" advertisers that you don't like. A real-life example is the word "evolution". Fundamentalist religious outfits have paid for that word and consequently you see their ads all over biology and scientific sites that mention it. I click on the links since the concept of religious crazies paying scientists is deliciously ironic. This too could classify as "click fraud".
Until the day that Google installs mind probes in every human being, it seems unlikely that they can do anything about either of these common situations. As long as such "click fraud" is essentially random and indistinguishable from background noise there is nothing they could do to stop it. Nothing at all.