|Take adsense off worst performing pages|
and get higher adsense revenue?
| 8:22 pm on Dec 2, 2005 (gmt 0)|
My worst performing pages are significantly worse than the average page on my site, but they do bring in a few $. The worst perfoming pages have:
- 40% of the CTR and ePCM of the site's average
- 60% of site's page impressions
- Bring in 25% of site's adsense revenue
I have heard that if I remove adsense from those pages, I am likely to see adsense revenue rise from the rest of the site. I do not understand how this could be, perhaps it is something to do with the mysterious smart pricing?
In your experience, is it likely to be worth removing adsense from the worst performing pages? How long would I need to run such an experiment for?
| 8:38 pm on Dec 2, 2005 (gmt 0)|
IMO, those pages are earning too much for this to work. I have done this, but only with pages that get either no clicks at all or very few, and that don't contribute to the site's earnings. 25% of earnings is too much to risk. A few percent would be OK.
Also, have you done everything you can to improve performance? A low CTR can be improved.
IF this process works, and it doesn't on all sites, it works because you increase your site's CTR and get a higher earning per click as smart pricing adjusts.
Many say smart pricing does not take CTR into account. That may be, but it does seem to take something RELATED to CTR into account.
Try to identify the pages that are getting no clicks or very few. They may only be getting 10-20% of overall impressions. But taking the code off can still have an impact. Try it for a week or two. If you see no change in overall earnings, you might as well put the code back.
| 1:19 am on Dec 3, 2005 (gmt 0)|
hunderdown, nice advise. is there a way how can we see which pages are not performing well? i have tried it using channels but it shows me only the directory and not the pages itself. as of now my highest CTR is 12.8, is this very low?
whats the best way to monitor CTR, is it by directories, domain or per adblocks?
| 2:02 am on Dec 3, 2005 (gmt 0)|
Just took adsense off some slow to no pages and wrote all data down to see if it will help out.
Keep you posted.
| 3:56 am on Dec 3, 2005 (gmt 0)|
digic, you must be using URL channels?
I use the other kind. This may not work for you--from your post I'm guessing you have a number of sites with a lot of pages. I give groups of pages that I feel belong together the same channel code, either because they are on the same topic or have the same set-up, such as review pages. Some channels are just one page. From that I can start to see what pages are working and where I either need to make improvements or put something in place of AdSense. To keep things simple, when I conclude that a particular set of pages is working well I put them in a catchall channel of pages I can leave alone--this helps to reduce the total number of channels.
I also don't tend to have more than one adblock per page, so I don't look at adblocks. I think if I had a site where multiple adblocks might work, I'd study them first, then starting tweaking pages....
But I've only got a few hundred pages to deal with. It might not be worth working at this level of detail for you.
Re a CTR of 12.8. That's pretty good! I seldom have a channel with a CTR that high. Mine is an old, informational site, pre-AdSense, so that's probably why.
| 4:43 am on Dec 3, 2005 (gmt 0)|
hunderdown, i am tracking both url channels and ads channels.
actually i am just optimizing one website. it might work for me, renaming my channels as per group, topic, or category might track and do changes from there.
sorry, my 12.8 is an average page ctr. ads/link channels range from 3 to 14.3
my site is just a 3month old, not so popular and im still tweaking it to get more visitors, high ctr and more revenue.
thanks a lot for sharing your ideas.
| 9:16 pm on Dec 3, 2005 (gmt 0)|
Thanks for your advice. You noted:
"I have done this, but only with pages that get either no clicks at all or very few, and that don't contribute to the site's earnings."
Did this result in a noticeable change in revenue for your site?
Ann, I look forward to hearing how removing adsense from some of your pages works out.
| 9:34 pm on Dec 3, 2005 (gmt 0)|
Guess I don't understand the logic here, sorry.
| 2:12 am on Dec 4, 2005 (gmt 0)|
The theory is that if you remove the worst performing pages (those with worst click through rates, worst epcm etc), then smart pricing detects you have a better performing site for placing ads on and rewards you by paying you more per click across the rest of the site.
That is the theory I have heard. I wanted to know if there is any truth in it and if anyone has actually tried it on their site and successfully raised their revenue.
| 7:45 pm on Dec 4, 2005 (gmt 0)|
I have done this on three separate occasions successfully, and once unsuccessfully. On the three successful occasions, the pages that I took the code off had 10% or more of total impressions for the site. Earnings then went up at least 10% and stayed up. The increase was too large to be a random fluctuation. The time that nothing happened, I removed code from a few pages that only contributed a few % of the total impressions.
I can't say this will work for every site, but it did work for me.