How Exactly Does Google DEFINE a Fraudulent Click?

Google has had an excellent PR day with the announcement that it will disclose to advertisers how many fraudulent clicks are being caught in Google’s filter. Yes, this is step in the right direction, and all that flowers and sunshine, but I have one question:

What EXACTLY is the definition of a fraudulent click?

In his independent report on click fraud, Alexander Tuzhilin puts his finger on the problem — Google can’t disclose their definition of click fraud:

An operational definition cannot be fully disclosed to the general public because of the concerns that unethical users will take advantage of it, which may lead to a massive click fraud. However, if it is not disclosed, advertisers cannot verify or even dispute why they have been charged for certain clicks.

Google is hoping that advertisers will be placated by having SOME transparency after so much obfuscation — and advertisers certainly have a lot more information than they did before — although anything more than NOTHING is going to feel like a lot. But advertisers are still looking at click fraud through a glass darkly, and they are still essentially at the mercy of Google’s system.

This disclosure of data doesn’t solve what Tuzhilin called the “fundamental problem of invalid (fraudulent) clicks.” All Google can do is try to keep the system afloat until they can transition to a cost-per-action model that doesn’t have the same vulnerability — and try to do so without tanking their stock price.