Comments on: Click Fraud, Google AdWords and gclid

by: Mark

Thu, 08 May 2008 09:00:13 +0000

Hi All,

I’m pretty new to this search marketing, so forgive me if this is a silly question.

I use affiliate marketing to generate revenue for my web site (Getting paid when per sale for Experian & Equifax credit reports etc.).

One of the affiliate marketing companies now captures all referall details (Yahoo / Overture id’s, keywords etc. & Googles gclid).

With the Yahoo information, I can easily identify which keywords have successfully converted for me.

Is there a way for me to find out which particular ppc keyword the gclid actually relates to in the first instance?

This would really help me to work smarter with my ppc budget, and evaluate where my profit is actually coming from

Many Thanks,
Mark

by: Dave Lineman / ClickTrue.net

Tue, 06 May 2008 20:46:26 +0000

Too add to this thread about click quality and click fraud, our company has been looking at this for many months as well. Advertisers are seeing an overall increase in volume and lowering of quality across the board. What we are seeing is that the obvious “click farms” are giving way to a large group of sites on the “official” partner networks from Google and Yahoo. We have a system that immediately “crawls” a web site as soon as you receive a click from it. While this won’t eliminate click fraud, it allows people to almost immediately detect suspect clicks and quickly block these sites. Web 2.0 style voting allows advertisers to benefit from the voting and blocking of other advertisers. Until Google and Yahoo dramatically increase their scrutiny of new sites, this is the best way we have seen to help stem the tide.

by: Jeremy Chatfield

Fri, 02 May 2008 09:38:24 +0000

@Nancy - IME, this affects a small fraction of users. It is findable, and fixable, by Google. It is worrying that autotagging does seem to be enabled without positive action, and that automatically enabling autotagging without a check by Google that it is safe, can destroy any value that you might have gained from advertising.

I’m reasonably certain that automatically enabling content match, “edgy content” pages and low quality domain parks accounts for more profit to Google than auto-enabling autotags.

Since you can’t use gclid tags, and you probably don’t have any other tags on the destination URL, you won’t be able to positively identify whether visitors arrive from Google AdWords - when you have Content Match enabled you may see a large fraction of visitors arrive from sites that are not Google owned. Given that you have had gclid problems, you may well see reduced visitor counts.

If you still see reduced counts after disabling autotagging then there’s probably a different problem - Google is usually *very* good about delivering paid clicks. The main question with Google should be the quality of the click, rather than the volume. I have cross verified hundreds of thousands, perhaps millions of clicks over the years, and delivery is simply not a significant problem.

by: Nancy

Sat, 26 Apr 2008 19:11:44 +0000

Additional note –I forgot to mention that both my hit counter and my stats at 123count.com show fewer than 1/4 of the hits (on just one of my landing pages) than google has charged me for since I put that page up.

by: Nancy

Sat, 26 Apr 2008 19:02:30 +0000

I found this page when searching for the “gclid” code info. I was doing his because I realized that for weeks now, I have had no business from my google ppc advertising and yet they are charging me more than ever. When looking at some server info from my backroom at bluehost, I discovered that people were receiving a 404 page not found message and the links they were going to were my website link with the gclid code tacked on to the end. So then I did searches on google that pulled up my ppc ads and when I clicked on them, sure enough–page not found and the links in the address bar had that code tacked onto the end. I won’t bore you with all the details except to say that altho Google is quick to suspend my ads when my website is down, they clearly don’t notice when their own codes debilitate the link. After finding your article I discovered the feature of autotagging (which I had never heard of) and went into my google adwords account and sure enough it had been enabled. I disabled it and now the links are working fine. But, Google owes me hundreds of dollars…and I didn’t set that autotag to on…so who did? I recently read a NY Times report on Googles high profits.. could this be why? Because my business had tanked..I went in and added new search words and upped my bids etc. etc…..all the while, people were clicking on my ppc ads and going nowhere. This is not good.

by: Jeremy Chatfield

Fri, 08 Feb 2008 09:06:46 +0000

Hi John - the “anonymous” comment was mine - it’s a pingback from a more recent article here. I’ll write a more comprehensive article about the problems that I ran into when I tried something similar, two or three years ago. Perhaps you’ve solved them :)

by: John Nagle

Fri, 08 Feb 2008 05:12:17 +0000

The anonymous comment above “while making new, ranked, spam sites is faster than detecting and removing them … the problem will continue to drag down all content advertising” deserves an answer.

The technical problems of blocking the bottom feeder ad sites quickly can be solved. SiteTruth goes out and looks at any site on request, and comes back with a rating immediately for ones it knows about. New sites take a minute or so to rate. So it is technically possible to keep up with the domainers.

The conventional wisdom is that this problem can’t be solved. The conventional wisdom is wrong.

John Nagle / SiteTruth

by: Spam in Comments, Unattributed Content | Merjis Internet Marketing Blog

Fri, 08 Feb 2008 03:59:10 +0000

[…] An older article has an interesting and recent comment by John Nagle, about a system to reduce exposure to spammy sites. Personally, I don’t think it will work for most advertisers. IME, the sheer quantity of low volume sites means that such a system will expend a lot of effort to deny sites that wouldn’t re-appear anyway. Put it like this… In a few tens of minutes I can sign up to new free domain host, and be publishing a spammy site, with an aged domain. However, getting traffic from each advertiser, sufficient to trigger a rejection, is a slow process - the site will make money. I believe that while making new, ranked, spam sites is faster than detecting and removing them (for all advertisers - not just the one) the problem will continue to drag down all content advertising. […]

by: John Nagle

Thu, 07 Feb 2008 20:27:53 +0000

Advertisers here are complaining about “low-value clicks” draining their ad budgets. Typically these are from “bottom-feeder” web sites made for advertising. There’s interest here in trying to generate better “exclusion lists” for Google AdWords, but nobody seems to have anything that works.

We may have something.

We have a system, available at SiteTruth.com, which evaluates web sites for business legitimacy. SiteTruth tries to find the real-world business behind the web site, reading through the site, checking business directories, checking SSL certs, and performing other tests. We’re working on this as a technology for improving search, and a patent is pending.

The online advertising community may be able to use these ratings, by extracting the referring site from incoming clicks, using SiteTruth to identify low-value domains, then blocking them via the Google AdWords exclusion list.

So we’d like to run a test. If you are running an AdWords campaign with high per-click costs, know which clicks actually generate revenue, and have the tools to extract the referring domain from your clicks, we’d like to talk to you. We’ll want you to send us a list of referring domains (up to 10,000 or so). We’ll rate them and send the ratings back. Recalculate your ROI with the bottom-feeder domains excluded from both ad cost and revenue, and tell us the results.

John Nagle / SiteTruth

(Data available for US and UK only, please; we only have business databases for a few countries.)

by: Jeremy Chatfield

Tue, 15 Jan 2008 08:20:37 +0000

Hi Nicki C -

covered briefly in the tenth paragraph of the article. IME, this affects people using web stores with “primitive” search queries embedded in page URLs, who are directing visitors to a page in the web store. From my client base, it’s around 1% of “normal” small advertisers. That’s still in the area of few hundred million dollars since this was introduced - I’ll hazard the guess that there could be a US class action suit in this observation.

My fifth recommendation would solve this - if Google added a dummy gclid when auto-tagging is enabled, then Google’s bots, Editorial Reviews and the account holder clicking on adverts, could all see the problem. Diagnosing the cause of a 404 with a valid base URL would be somewhat tricky and may be why Google hasn’t done this, considering how easy it should be, technologically.

Cheers, JeremyC.