Using CSS to increase your click trough

I have use the css div, in my one of my old site that getting 1K unique per day and almost triple my earnings... so I decided create another site with the same format.... here is the sample site that I did

http://www.anger-management.be/

I put all the codes at last part of the source code.... then use the div absolute properties to position them in the place I want. The reason why I put them at the bottom is that I will allow all content fo the site be displayed first before the ads, this will ensure that the robot will give related ads.

Try it.... hope may help

 

wow.......that's spammy as hell

 

It's a sample... just made that to show how I make my other site... you can change the content... but the layout I guess help to increase the CTR.

 

Is this right? I never had the impression that the physical location mattered to the ad server - doesn't it come back and read the whole page anyway?

A typical log entry:

66.249.66.8 - - [15/Jan/2006:10:12:36 +0000] "GET /Unixart/perl_eval.html HTTP/1.1" 200 12436 "-" "Mediapartners-Google/2.1"

Code (markup):

shows Google reading the whole page. Or am I misunderstanding something?

 

I guess the position of the ad is always matter.... because there are lot of people getting problem... displaying ads on their sites.

 

Are we talking about the same thing here? Of course the visible position matters, but your post implies that the position of the ad code in the page matters.

Forgetting CSS for a moment, even a simple table layout means that the physical position of the code can be different from what a visitor experiences (just mentioning that for the benefit of those who didn't understand what the css does). So your post implies that when the robot comes to read the page, it's going to pay attention to the text that precedes the ad?

Yet if that were true, wouldn't the ad code have to come after the section targetting tags? I see nothing that implies that.

I don't think this is true. I think the robot reads the whole page and decides what the ad should be from that (including section targetting of course).

But if you can point me at anything that says otherwise, I'd be interested to read it (and would be frantically adjusting layouts immediately).

 

I don't think there is any relation between the relevancy of ads displayed and the position of the ad codes. Where ad code position does come into play is when you have more than one ad units on the same page.

Adsense will place the highest paying ads in the ad unit that appears first in the page code. That's why you have to make sure that the ad unit that gets the highest CTR is the one that occurs first in your page code.

 

Right.

But here's a question that's related to all this:

I make a post, and read it myself so that I can see how it looks in different browsers, etc. I could do this off-line before I post it, but there's so much dynamic stuff in my pages I like to see "reality". So.. first time it loads, a read by Google MediaPartners shows up in the logs. I assume that's the ad server reading the post to analyze it for relevancy to its ad inventory.

Now.. I look at the post and say "Aw shoot, I left out a bunch of important stuff" and drop back into my editor (vi, of course) and add a couple of new paragraphs.

Or I see the ads and I say "Ooops - Google is misinterpreting that page" so I rewrite it and sprinkle the "right" keywords and maybe use section tags etc. in hopes of helping Adsense be more relevant.

When will the ad server re-read the page? Looking at my logs, I see no clear pattern to that -sometimes it reads again within seconds, but sometimes not for quite a while. For example, here are recent logs of a page that typically gets a fair amount of pageviews a week:

66.94.237.142 - - [06/Jan/2006:12:58:17 +0000] "HEAD /Basics/sudo.html HTTP/1.0" 200 - "-" "Mozilla/4.0 (compatible; MSIE 6.0; Windows NT 4.0), Mediapartners-Google/2.1"
66.94.237.140 - - [06/Jan/2006:12:58:17 +0000] "GET /Basics/sudo.html HTTP/1.0" 200 26315 "-" "Mozilla/4.0 (compatible; MSIE 6.0; Windows NT 4.0), Mediapartners-Google/2.1"
66.94.237.141 - - [06/Jan/2006:12:58:17 +0000] "HEAD /Basics/sudo.html HTTP/1.0" 200 - "-" "Mozilla/4.0 (compatible; MSIE 6.0; Windows NT 4.0), Mediapartners-Google/2.1"
66.94.237.142 - - [06/Jan/2006:12:58:17 +0000] "HEAD /Basics/sudo.html HTTP/1.0" 200 - "-" "Mozilla/4.0 (compatible; MSIE 6.0; Windows NT 4.0), Mediapartners-Google/2.1"
66.94.237.140 - - [06/Jan/2006:12:58:18 +0000] "GET /Basics/sudo.html HTTP/1.0" 200 26089 "-" "Mozilla/4.0 (compatible; MSIE 6.0; Windows NT 4.0), Mediapartners-Google/2.1"
66.94.237.142 - - [06/Jan/2006:12:58:18 +0000] "GET /Basics/sudo.html HTTP/1.0" 200 26089 "-" "Mozilla/4.0 (compatible; MSIE 6.0; Windows NT 4.0), Mediapartners-Google/2.1"
66.249.66.8 - - [11/Jan/2006:02:57:26 +0000] "GET /Basics/sudo.html HTTP/1.1" 200 26382 "-" "Mediapartners-Google/2.1"
66.249.66.8 - - [13/Jan/2006:03:50:54 +0000] "GET /Basics/sudo.html HTTP/1.1" 200 26384 "-" "Mediapartners-Google/2.1"

Code (markup):

Notice that on the 6th it re-read (or at least checked with HEAD) several times, but then didn't look again till the 13th. In that same period, that page logged over 1500 pageviews so obviously the ad server had plenty of chances to re-read - though I can't easily tell how many of those were clients actually capable of triggering the ad code. I'd assume it was considerably more than the few re-reads indicated above (and I could check with Asrep if I cared).

 

It's difficult to say how often the page gets revisited. I think the bot checks if the page has been modified or not. However it must be coupled to the load as well. Seeing as sometimes it takes a few minutes to show relevant ads and other times a few hours or even a couple of days, after uploading a new page.

 

Not sure if it is related to your technique, but I get tons of javascript errors when trying to browse your articles.

 

The only way it could check for modified is through a HEAD - which it does do sometimes, but as you can clearly see in the logs I posted, it apparently sometimes just does a GET - which implies that it doesn't have the page data in its cache at that time.

Probably there are multiple ad servers and they may not share the same cache.. and not all them may coveniently identify themselves as MediaPartners, so I may not be seeing all the accesses either..

 

That must be the case. In any case I was only speculating on a possible way the Adsense bot acts. I've never really gotten round to trying to identify exactly how Adsense does what it does.

 

And if we ever figured it out, they'd change it immediately

 

Never a truer word was spoken! We'll be playing the guessing game forever if Google get their way.

 

RSS is great with Adsense. Use it all the time and increased my earnings by 15 percent.

 

CSS not RSS

 

Another board spammer who doesnt even read the full thread

Back to what I was going to say.

I would see it as more of "in your face" rather than spam, lol.


lakay - does this method work well?

 

Am I seeing the page right? I am counting for adsense units on the page? I thought the limit was three? Is that becuase one of them is a link unit?

Thanks
Ichibanda

 

Yeah three Ad blocks and the link unit in the Yellow/Orange Nav at the top.

 

Cool! I always thought the link unit counted as one of the 3 allowed ads. Time to add some link units to a few of my sites.

Thanks!
Ichibanda

(off to add link units!)