I was getting frustrated because I was getting a large number of clicks, but low payouts across my sites. Right now, I have a few sites going, and I know which ones have the best paying clicks. Unfortunately, some of my more 'popular' sites seemed to be dragging down my payout.
To increase my earnings, I wrote an ad-rotator system to switch between CPA and contextual ads. I was playing with different settings trying to maximize the monetization, when something really interesting happened.
I was tweaking the CPA settings, and I had them set to display 70% of the time on my highest traffic site. I got tired and went to bed with the settings in place.
When I got up in the morning, I checked my adsense, and my clicks were paying out MUCH higher. I looked at my channels, and my highest traffic site had the lowest number of clickthroughs. Because I had the CPA's showing 70% of the time, the adsense on that site showed 70% less.
My other sites still had the same number of usual clicks (I was only experimenting on my high traffic site). But, the clicks were paying out much better. I turned the rotation up to 80% and tried it again. The next day, the clicks paid more AGAIN.
I have heard that smart pricing affects an entire account. And, I'm not sure how it works. But, it seemed to me that as I started reducing the visibility of the Adsense on my high traffic site, the payout on my low traffic sites got bigger.
I'm not sure if this is smart pricing, or if there is an algorithm that tries to maximize profits for google by estimating the max potential clicks for a site, then reducing the payout accordingly. Either way, i've been getting 30% more per click since this experiment started.
The bonus? My CPA ads are starting to pay out more than the adsense used to. So, my main site is generating more $$$ through CPA, and the others are getting more per click than before.
It seems like this would be an easy problem for an optimization algorithm. You would just have to get the channel data and the cpa payout info, then automatically tweak the display percentages across your sites and measure the differences.
I think I'll build that kind of optimization into my next generation of automated sites. But for now, I thought I would share my findings. If you have a network of sites, and you notice lower trending payouts, try isolating the junk-clicks and reduce the amount of adsense you display on that site.
When I did this, my other sites started paying out much better.
It can't be the rotation, because the ads that are getting clicked are on completely different sites. Suppose there were 2 sites...
Site A -> 100 clicks -> $1
Site B -> 10 clicks -> $1
Now, I reduce the number of impressions on Site A by 70%. This is what it ends up doing...
Site A -> 30 clicks -> $0.30
Site B -> 10 clicks -> $2.00
Site B still has the same number of clicks, but each click is worth more $$ now and the overall payout is GREATER by serving less ads.
I thought that maybe it was just fluctuation in the advertising, but I don't think so. The obvious way to test it is to increase the number of impressions I'm serving on site A and see if the payout goes down again. If I do that, I'll let you all know how it goes.
Those ads are still showing up with the same frequency, and they are still generating the same number of clicks. It's just that the clicks on those sites are worth way more $$$ now.
To increase my earnings, I wrote an ad-rotator system to switch between CPA and contextual ads. I was playing with different settings trying to maximize the monetization, when something really interesting happened.
I was tweaking the CPA settings, and I had them set to display 70% of the time on my highest traffic site. I got tired and went to bed with the settings in place.
When I got up in the morning, I checked my adsense, and my clicks were paying out MUCH higher. I looked at my channels, and my highest traffic site had the lowest number of clickthroughs. Because I had the CPA's showing 70% of the time, the adsense on that site showed 70% less.
My other sites still had the same number of usual clicks (I was only experimenting on my high traffic site). But, the clicks were paying out much better. I turned the rotation up to 80% and tried it again. The next day, the clicks paid more AGAIN.
I have heard that smart pricing affects an entire account. And, I'm not sure how it works. But, it seemed to me that as I started reducing the visibility of the Adsense on my high traffic site, the payout on my low traffic sites got bigger.
I'm not sure if this is smart pricing, or if there is an algorithm that tries to maximize profits for google by estimating the max potential clicks for a site, then reducing the payout accordingly. Either way, i've been getting 30% more per click since this experiment started.
The bonus? My CPA ads are starting to pay out more than the adsense used to. So, my main site is generating more $$$ through CPA, and the others are getting more per click than before.
It seems like this would be an easy problem for an optimization algorithm. You would just have to get the channel data and the cpa payout info, then automatically tweak the display percentages across your sites and measure the differences.
I think I'll build that kind of optimization into my next generation of automated sites. But for now, I thought I would share my findings. If you have a network of sites, and you notice lower trending payouts, try isolating the junk-clicks and reduce the amount of adsense you display on that site.
When I did this, my other sites started paying out much better.
It can't be the rotation, because the ads that are getting clicked are on completely different sites. Suppose there were 2 sites...
Site A -> 100 clicks -> $1
Site B -> 10 clicks -> $1
Now, I reduce the number of impressions on Site A by 70%. This is what it ends up doing...
Site A -> 30 clicks -> $0.30
Site B -> 10 clicks -> $2.00
Site B still has the same number of clicks, but each click is worth more $$ now and the overall payout is GREATER by serving less ads.
I thought that maybe it was just fluctuation in the advertising, but I don't think so. The obvious way to test it is to increase the number of impressions I'm serving on site A and see if the payout goes down again. If I do that, I'll let you all know how it goes.
Those ads are still showing up with the same frequency, and they are still generating the same number of clicks. It's just that the clicks on those sites are worth way more $$$ now.
No comments:
Post a Comment