Forum Moderators: martinibuster
It declined from 28.3% in the fourth quarter of 2007 but increased from 21.9% in the first quarter of 2007.
You can read the press release on the Click Forensics website.
I can't vouch for the accuracy of their data, but I know that they have been around for awhile and work with a lot of large companies including Yahoo and Lycos. I know that Google and Click Forensics have had some disagreements. Obviously, Google does not like to see the publication of statistics such as these. They are not exactly reassuring for advertisers.
First of all, if you know the rate of click "fraud"--and I would really like to know how they define that!--then you must have an idea of which clicks are "fraudulent." And that should mean you can just kick them out of the system. The advertisers don't get charged for them, and everyone's happy.
Even if you can't pinpoint them, you can make adjustments, through mechanisms like smart pricing.
No. Just lower your bids--and that's assuming that these fraudulent clicks aren't detected. Advertisers know that many clicks don't convert. A fraudulent click (and again, what does that mean?) is just another kind of non-converting click.
Like shop-lifting or employee pilfering from bricks and mortar stores, you plan on it.
Frankly, it's not the advertisers who should be upset by this stat (IF it's even accurate...)--it's the honest publishers. A fraudulent click is money that they aren't earning.
As far as Click Forensics is concerned, how would they even know this measure? This is pure rubish. Only Google and Yahoo can guess - and they don't even know the real answer.
Nope these guys are just shooting a number out to get attention but really that is scary no one really knows so any number posted could be just a number an nothing else or it could be right on. That's the problem.
Comscore, a larger and more authoritative company lacking conflicts of interest did a click report analysis and it was still way off. I'm having trouble believing anyone's guesstimates, whether they're praising Google or whacking Google.
Advertisers pay millions to have a 30-second advertisement run during the Super Bowl. Why? Because it is watched by millions of viewers. What would happen if everyone changed the channel during that commercial? Should they get their money back? Did the networks defraud the advertiser? Do they do anything to try to prevent people from changing channels? Are they overstating their actual audience?
Preventing fraud maintains the integrity of the entire system. What determines a bid is based on economics, not fraud. For example:
Let's say I know my net margin is $100 on every sale and my conversion rate with Google is 10%. This means I can pay up to $10/click and still break even.
But let's say fraud starts to infiltrate the network to the tune of 50% of all clicks.
Simple math tells me that my net margin is still $100 but my conversion rate is now 5%. Now I can afford to pay only $5 per click and still break even.
And as a publisher that runs a clean site, I want Google to maintain a constant watch over fraud because it hurts my bottom line.
I don't deny that fraud is a problem, but I'm just not convinced it's a huge problem. Yeah, if 50% of the clicks are fraud (or 27% as the report claims) AND if Google doesn't detect any of them AND if there isn't smart pricing or some other discount scheme in place, then it's a big problem.
But the click fraud rate isn't that high, Google does detect at least some of the bad clicks, and there is smart pricing. So it's not that big a problem....
And it does hurt our bottom line, as I pointed out earlier. But by how much?