Higher impressions lead to lower CPM

That's the trend we saw!

[fokfok]

A month ago, we set up 2 channels, we gave 25% of the hits to channel A and 75% to channel B. After 2 weeks, we adjusted the ratio, so 50% go to channel A and 50% channel B. Before the switch, channel A constantly earned a higher CPM than channel B. But after the switch, channel B constantly outperformed channel A. Obviously, if you are just comparing two channels, channel B is more lucrative. But this test also reveals a possibility that Google pays lower CPM to larger publisher. What do you think?

If there are any large publishers out there in this forum ($500+ daily adsense income), would you mind to tell me a vague range of eCPM you are getting? You know google don't want us to reveal the exact eCPM...

[ogletree]

That test is extremely flawed. CPM has no bearing on what you get paid it is just a stat to help you compare things. If you have the same amount of traffic and change your channels to get higher CPM you still get the same as you would have if you did not have the channels. Also if you notice that your channels do not show exact numbers. Add up the totals for your channels one day they don't add up to the total amount you earned that day. Plus there are so many factors that go into what you get paid you have no idea if your data is correct. There is no way to really run tests on adsense there are just too many factors and several of them are completely unknown. Especially when one click can be $0.01 and the next be for $1.05 from the same page. They keep it this way on purpose. If you are getting a high CPM like $200 and you can get a lot of new traffic that you know or find out lowers that to $50 but you make more money per day don't throw away the traffic.

The only thing you need to do to get more money is:
1. Get more visitors
2. Get more targeted ads
3. Have content that gets higher paying ads
4. Ad position
5. look of the ad
6. Have content that gets the visitor interested in the subject of the ads so they want to get more info or buy what the advertiser is selling or have content that does not provide any help at all so the visitor has to click on the ad to get more info.
7. Get visitors that are looking for what your site is about.

These are not in any order.

[GeeWhizzler]

Larger publishers are on a different contract that the standard contract. What the terms of those deals are can't be tested for and you are wasting your time, imho.

[fokfok]

Well, we were instructed by google's rep to set up those channels. I'll post some stats for you guys.

[fokfok]

ogletree says:

That test is extremely flawed. CPM has no bearing on what you get paid it is just a stat to help you compare things. If you have the same amount of traffic and change your channels to get higher CPM you still get the same as you would have if you did not have the channels. Also if you notice that your channels do not show exact numbers. Add up the totals for your channels one day they don't add up to the total amount you earned that day.

Sorry i was not clear enough. Channel A and B are two channels that appear on the same page. However, channel A targets to the original page while channel B targets to another page. So in this test, I was trying to find out which kind of targeted ads was more lucrative.

But well, I guess google has different ads on different days for channel A and B, may be it just happened that google was delivering some better ads to channel B after the switch. However, the result was pretty consistant so far, and the eCPM change was significant. That's why I suspect may be it's not something random.

[GuruSense]

Here is my answer i reach 400-500$ and immidiately I see a fall ... it's called "you ran up your advertiser now you must wait until the next day".

simple