Do adsense blocks need time to "settle in"

[yankees1976]

On my blog I have 2 300x250 blocks blended into the content, and one 160x600 in the side bar.

I was getting decent CTR for both, but the 300x250 blocks were doing noticeable better. I decided to take out the 160x600 side bar block to see if it would force more relevant and higher paying ads into the 300x250 blocks, but right after doing so, my CTR dropped drastically to almost 0%, and the revenue per click dropped by about 90%.

Is this because the ads that were normally being served to these two blocks became optimized, and now need time to reset and gain traction again, or is it just an unfortunate coincidence.

I also changed hosts in this time period, but there was no down time, there are no broken links, and no other disruption to speak of.

[dburk]

Hi yankees1976,

Removing that one ad block is not a likely cause for such a sharp drop in CTR, unless you significantly changed the layout or somehow broke the page code. I would look for issues that the googlebots may be having crawling your website.

I have seen issues with some web hosting companies "accidentally" blocking search engine spiders. You need to consider security, has anyone gotten access to your web page code and replaced your publisher ID with their own? Have you inspected the ads displaying on your pages to see if they are PSA or otherwise irrelevant.

Have you checked the diagnostic section of your AdSense account?

p.s. Was there any down time while transferring web hosting companies? That would do it.

[yankees1976]

Originally Posted by dburk

Hi yankees1976,

Removing that one ad block is not a likely cause for such a sharp drop in CTR, unless you significantly changed the layout or somehow broke the page code. I would look for issues that the googlebots may be having crawling your website.

I have seen issues with some web hosting companies "accidentally" blocking search engine spiders. You need to consider security, has anyone gotten access to your web page code and replaced your publisher ID with their own? Have you inspected the ads displaying on your pages to see if they are PSA or otherwise irrelevant.

Have you checked the diagnostic section of your AdSense account?

p.s. Was there any down time while transferring web hosting companies? That would do it.

dburk,

Thanks for the quick and detailed response. Let me see if I can address those questions & comments in order:

-Definitely no changes to code on the site, I checked the widget areas for that. I did plug in some CPM square ads to see how they would do (only 2) but those got about 4 clicks from 1000s of page views.

-As for the google crawling, Ive got a small handfull of things that come up using the Google Site Diagnostics tool, but they have been there since well before I had any adsense success.

-What kind of question could I ask my host regarding an accidental blocking crawlers? I know the site has been crawled recently though, Nov. 8th was the last time I'm able to verify, and the host transfer was completed on Nov 2nd.

-As far as I know, nobody else has access to the admin portion of the site. My adsense ID is still accurately input in the code. Even if it was wrong though, I see pretty much all of the adsense related exit links from my site using statcounter.com, but there just are not as many there as there used to be.

-No PSA ads that I've seen, all of the ads look very familiar compared to what I used to see before. I've also used the competitive ad filter to remove all the junk ads, did that a few months ago.

-Is there anything I might be missing in the diagnostics tool besides the links that couldnt be accessed during the crawl?

-There was a total of about 3 minutes down time during the transfer when they needed to reboot a server, and thats it.

Again, thanks for your response and detailed troubleshooting efforts...still scratching my head here so if you have any other thoughts it would be much appreciated.

[dburk]

Hi yankees1976,

When you move from one host to another there is a period of about 72 hours that some of your traffic could be going to the old IP adress. To have a smooth transition you need to have your website up and running at both hosts, the old and the new. 72 hours after you update your name server you can safely take down the old website. If you take it down too soon you could create a problem if traffic goes to a page that can longer be found at the old IP address.

Obviously you need to fix those issues where links cannot be found. If the bots encounter too many missing links in a row they will give up and comeback later. If they comeback and find the same issue they will wait even longer before returning.

[yankees1976]

Originally Posted by dburk

Hi yankees1976,

When you move from one host to another there is a period of about 72 hours that some of your traffic could be going to the old IP adress. To have a smooth transition you need to have your website up and running at both hosts, the old and the new. 72 hours after you update your name server you can safely take down the old website. If you take it down too soon you could create a problem if traffic goes to a page that can longer be found at the old IP address.

Obviously you need to fix those issues where links cannot be found. If the bots encounter too many missing links in a row they will give up and comeback later. If they comeback and find the same issue they will wait even longer before returning.

Thanks again...there was a 48 hr period of time that both the old host and new host were active, but there was never a drop off in unique visitors, or page views, in fact, I managed to show increases in both over that period of time. Any other thoughts?

I do need to fix the issue with the links, but as I mentioned those have been there for a very long time, including when the adsense was doing well.

[dburk]

Hi yankees1976,

Okay, it's beginning to sound like a traffic source problem.

If you are using a blog you can sometimes suffer from a relevancy issue. Because the front page of a blog shows recent blog posts you could build traffic targeting something on the front page that eventually disappears from front page as new posts push it off. Since the traffic targeting the previous posts do not find the page or the ads relevant they hit the back button.

If you have a good traffic stats program, this will show as a high abandonment rate for the homepage. Of course if they leave by clicking an ad that's good. The problem is if the content isn't relevant to their interest then neither will the ads be.

[InternetMarketingIQ]

Removing ad blocks will NOT change relevancy it will simply remove ads. In theory this should reduce the chances for low performing ads to appear on your pages. CTR problems are on your end if the click rates go down.

The issue is with changing IP Addresses by changing hosts. Anytime you do this it will take Google a bit of time to recrawl.

[Fraggler]

Another way to help relevancy is by filtering out the irrelevant categories. I posted about it here Adsense Category Filter

I have seen an increase in from $88 eCPM to $110 eCPM (through CPC not CTR) since putting it in place. I have no way of knowing if this can be attributed to the filter but it has helped.

Google does take time before placing the expensive ads on your site by the way. They call this smart pricing. They will discount the costs of advertising on untested sites so it doesn't ruin the cost per conversion for the advertiser. When your site/ad block shows that it converts then they will increase the cost (reduce the discount) to advertise there.

[yankees1976]

Originally Posted by dburk

Hi yankees1976,

Okay, it's beginning to sound like a traffic source problem.

If you are using a blog you can sometimes suffer from a relevancy issue. Because the front page of a blog shows recent blog posts you could build traffic targeting something on the front page that eventually disappears from front page as new posts push it off. Since the traffic targeting the previous posts do not find the page or the ads relevant they hit the back button.

If you have a good traffic stats program, this will show as a high abandonment rate for the homepage. Of course if they leave by clicking an ad that's good. The problem is if the content isn't relevant to their interest then neither will the ads be.

I guess the relevancy thing could be an issue, but honestly to have a solid month of what I consider a good CTR suddenly drop by 90% in the first week of the next month doesnt seem like it would just be related to relevancy. And with statcounter.com it tracks the key word searches by %, and my top 10 that I really monitor have switched places here and there, but thats always been the case and its nothing significant really.

Thanks again though, this is forcing me to think out every possibility and talk it through...eventually something will make sense.

[yankees1976]

Originally Posted by InternetMarketingIQ

Removing ad blocks will NOT change relevancy it will simply remove ads. In theory this should reduce the chances for low performing ads to appear on your pages. CTR problems are on your end if the click rates go down.

The issue is with changing IP Addresses by changing hosts. Anytime you do this it will take Google a bit of time to recrawl.

That was my intent to see if I could improve the performance of the remaining ads...clearly that has not worked out so well for me.

Im pretty sure that it is not related to an issue with the Google crawl. Google has crawled the site multiple times since the change. I usually had a once per day crawl from Google, in the 5-8AM range, now it seems to shifted back a little bit to like 4-7AM, but its still once a day.

Thank you for your feedback though...its all very helpful in trying to isolate what the issue could be.

[yankees1976]

Originally Posted by Fraggler

Another way to help relevancy is by filtering out the irrelevant categories. I posted about it here Adsense Category Filter

I have seen an increase in from $88 eCPM to $110 eCPM (through CPC not CTR) since putting it in place. I have no way of knowing if this can be attributed to the filter but it has helped.

Google does take time before placing the expensive ads on your site by the way. They call this smart pricing. They will discount the costs of advertising on untested sites so it doesn't ruin the cost per conversion for the advertiser. When your site/ad block shows that it converts then they will increase the cost (reduce the discount) to advertise there.

Awesome, I never noticed the category blocking. I just blocked 4 categories that had minimal or no earnings.

I used to get a decent $ per click on the ads, not anything crazy, but what Id consider reasonable...definitely not in that .01-.25 cent range that might make one think they were smart priced.

[yankees1976]

I just realized something that all of this back and forth reminded me about that might be the cause.

I was playing with a plugin to add page numbers to the bottom of the page, and when I did that I changed the #of posts on the front page from 10 to 5 to make sure it was working properly, but I never changed it back. Im assuming that could have had an impact on the number and variety of key words that the google ads had to work with, resulting in less targeting and fewer clicks? I just changed it back to 10 and Ill see how it goes over the next few days.

Thanks again for all of you for helping out, it was all helpful in forcing me to think about anything that could have changed.

[kwality]

it takes a little while for the adsense bots to display relevant ads. Longest I have had to wait is three days.

[euhlir]

I've noticed that even the slightest change of your ad blocks can have a drastic effect on the CTR rate of ads. That's why it's important to test out different ad sizes and layouts so you can find the best possible combination.

It does take about 15-30 minutes in some cases for the ads to be targeted more for your content, but after that it's live and should be considered active testing.