AdWords - Try Today's Challenge

[abbottsys]

Today's AdWords challenge question.
Level of difficulty: abbottsys testing ;-)

I copy an AdWords ad from a Google SRP and paste it into an (html) email so the ad is a live link. I send the email to a ton of people, and many of them click the ad in the email. Does AdWords track these clicks and does the advertiser get charged for them?

[searchengineman]

I would imagine the answer is no.

With out hard proof. I would imagine that there is a time stamp or identifier in the URL string, that lets Google know that Google served the AD. Otherwise an unscrupulous vendor would just email his competitors ads, for click fraud purposes. (Why not change the text as well!)

PS: That's a great question by the way.

Searchengineman.

[Discovery]

Asys, I had that concern at one time as well. Mostly because I would copy and paste an ad from time to time to send in email to a client or co-worker. I would modify the URL or remove it, just in case.

I believe the clicks get tracked, but the advertiser is not charged. How to test for this I'm not sure. Send it to somebody you know in a small town in Russia and see if it shows up in your geo overlay report?

Discovery

[Gooner151078]

I believe the ad would be disregarded. The same would be true if you copied the code of the ads to another HTML page. There must be something Google side that verifies the referring URL.

[abbottsys]

Answer: The first click is recorded and charged, all others are not.

I constructed a simple test. I ran a new ad and confirmed that it was getting no traffic. I then did a search on one of the ad keywords, found the ad, and copied it into an email. I sent the email to several people who clicked on the ad in the email.

AdWords recorded 1 click for the ad, in the search network, and on the actual keyword that I used to find the ad originally. All other clicks worked of course, but were not recorded or charged by AdWords.

The more interesting thing is knowing what mechanism Google uses to block this type of activity. I suspect they assign a *unique ID to every SRP they ever serve* and store this (or a reference to it) in the tracking code in the url. That way they have total control, knowing exactly which SRP generated a click.

With this method they could easily eliminate the above "manual SRP screen scaping" activity. However, this method would not allow them to eliminate more sophisticated (automated) screen scraping activities. But that could be the subject of another test :)