Breaking Out Of Frames Used By Google Image Search With Javascript

[jordanmcclements]

I get approx 50% of my Search Engine Traffic from Google Image search.

I had read at Don’t Lose Your Google Image Search Traffic - stefanjuhl.com (http://stefanjuhl.com/2007/dont-lose-image-search-traffic/) that it was worth trying a Javascript 'Frame Buster' piece of code to decrease bounce rate etc for visitors who find your site through Google Image search.

I was initially loathe to try it, as it affected the user experience (it 'sort of' disables the back button), and I also thought it may not be seen by Google as a good thing (about 75% of my traffic comes from Google!).

I started a thread on the google webmaster help forums : Breaking Out Of Frames Used By Google Image Search With Javascript - Crawling, indexing, and ranking | Google Groups (http://groups.google.com/group/Google_Webmaster_Help-Indexing/browse_thread/thread/60e3395dfeb9ab3c/e5296f30a6c40d8b)

and was persuaded to at least try it for a while to see how it went...

Here are my findings:-
It SERIOUSLY messes up your statistics in google analytics. ALL referral traffic from 'images.google' is now counted as 'direct' traffic.
It seems to count the initial page view from a google images visitor twice (I assume that the view has been registered a split seconds before the client executed javascript kicks in and reloads the same page again without the google image frame?) I suppose this is not a bad thing if you are trying to lie about the amount of page views your site gets every month!
My Adsense revenue was seriously damaged. I only let the web site run for 2 full days with the frame buster script installed, and I know that there can be fluctuations in adsense revenue, but the drop in earnings was much bigger than any drop I have ever encountered on my web site before - so it seems like too much of a coincidence for the 2 things not to be related. I am still not sure why this would be... Maybe when the page is loaded a second time, different, lower paying ads are shown? Maybe, ads are optimised in some way by google based on the referring search term which is then lost when the image search frame is removed?Has anyone else had any experiences with anything similar?

[dann]

Maybe, ads are optimized in some way by google based on the referring search term which is then lost when the image search frame is removed?
The fact that you have Analytics on the site anything they could use that data for they probably do. When I ran adsense on my sites referral data wasn't in the equation but it may be now.
The only other thing I can think of is the ads might be retrieved from Google twice doubling your impressions and cutting your CTR in half.
Are you getting less clicks? Or just less per click?
Personally though after reading the blog post that's not something I would do. All it takes is one complaint, and the fact that your analytics are getting messed up and adsense revenue is going down Google Bot might raise an eyebrow eventually. I've heard of people losing their adsense accounts for far less aggressive tactics.
My theory, good luck
Dan

[John]

:mad:
... affected the user experience (it 'sort of' disables the back button), and ...?

There are framebuster scripts that do not "disable the back button". See this page: Frames Buster - JavaScripts - Grizzly WebMaster (http://grizzlyweb.com/webmaster/javascripts/framesbuster.asp)

[jordanmcclements]

John, I have been put off all frame busters for life now anyway!

Dann, very roughly, I was getting lots of extra page views (probably caused by the javascript). After adjusting for the bogus extra page views, CTR was roughly the same, number of clicks was roughly the same, but revenue was much lower..

[Conficio]

Here are my findings:-
It SERIOUSLY messes up your statistics in google analytics. ALL referral traffic from 'images.google' is now counted as 'direct' traffic.
It seems to count the initial page view from a google images visitor twice (I assume that the view has been registered a split seconds before the client executed javascript kicks in and reloads the same page again without the google image frame?) I suppose this is not a bad thing if you are trying to lie about the amount of page views your site gets every month!

That is a function of how you install your frame buster script. You shoudl be able to surpress the Analytics script to run before the frame buster, so the first rendering is not counted.

My Adsense revenue was seriously damaged. I only let the web site run for 2 full days with the frame buster script installed, and I know that there can be fluctuations in adsense revenue, but the drop in earnings was much bigger than any drop I have ever encountered on my web site before - so it seems like too much of a coincidence for the 2 things not to be related. I am still not sure why this would be... Maybe when the page is loaded a second time, different, lower paying ads are shown? Maybe, ads are optimised in some way by google based on the referring search term which is then lost when the image search frame is removed?
See the comment above.

Your results sound to me caused by bad implementation and not by the fact it can't be cone.

K<o>

[jordanmcclements]

Your results sound to me caused by bad implementation and not by the fact it can't be cone.
Well, if anyone has managed to implemented it much better than I did - then I would be very grateful to hear exactly how they did it!

[Webnauts]

It SERIOUSLY messes up your statistics in google analytics. ALL referral traffic Did you try to implement this? Sulli's Google Analytics Tips and Tricks: Track Google Images Search in Google Analytics (http://blog.thesulli.net/2007/06/track-google-images-search-in-google.html)

[jordanmcclements]

Yes, I did implement that!

It worked really well (without the frame buster in place).

The only thing I have not worked about it is I have the following search terms appearing a lot under 'images.google':-

20
18
21

(and I have no idea why)..