When or how does CPM go up?

Does this have anything to do with how well your website performs?

[chopin2256]

Does CPM go up if Google sees that you are getting higher, more consistent and predictable numbers? For example, if Google sees that your site is bringing in many clicks, or has a high CTR, will Google raise the CPM, by giving you higher paying ads? I guess the main question is, will Google give the better ads to sites that are performing better versus the sites that are not really performing well?

Another question would be, does Google give higher payouts to performing sites that are starting to gain Google's trust?

[europeforvisitors]

Google isn't saying, so nobody here can answer either of your questions with more than a guess.

[ncw164x]

OK I will guess then

First question = No
Second question = No

There is a rumour that you get paid more per click when your account is new and as time goes on you start to get paid less and less per click, but it is only a rumour.

Bottom line is the amount of traffic to your site...more visitors = more clicks = bigger cheque...its that simple, just dont start using any of the schemes that drive 100,000 visitors to your site for $9.99

[Atomic]

There is a rumour that you get paid more per click when your account is new and as time goes on you start to get paid less and less per click, but it is only a rumour.

This is about as far from my experience as you can get. I started with pennies a click and now see well over a dollar a click. Of course, I am adding content all the time. Good, well written articles and lots of useful information are added regularly. So traffic is way up, too.

I can see Google paying less for click on forums and other sites where the clicks don't convert. But I just can't see a penalty for sticking with AdSense.

[ncw164x]

I see clicks at over $13.77 but that does not mean all my clicks are the same amount of money!

I did not say it was a penalty but there are quite a few people who I know have experienced the same but it could be relating to the type of site and the content on each page plus the smart pricing, put it all together and it "seems" like your earning less

[hunderdown]

There's some evidence that a very low CTR will actually reduce the EPC paid under smart pricing--to the extent that removing a poor-performing set of pages will actually increase overall earnings. Which is somewhat counter-intuitive, to have greater earnings for fewer impressions. But in discussions about this, some people have suggested that there might be other explanations for this apparent effect.

[DamonHD]

Hi,

Bar the normal Google noise, and a pleasant jump earlier this year, I have seen pretty steady CPM etc, with gradually rising earnings as my page hits rise.

I see no evidence for me of Google doing bait-and-switch on the EPC.

I'm adding new and diverse material all the time.

Rgds

Damon

[Visi]

This brings up an interesting question on how often "smart pricing" is updated for a site. Announced last April there was no schedule of reviews or timeframes. Now this one has me thinking:)

[MikeNoLastName]

My theory, and it's based only on stats on our own site (50K+ clickthrus/month), is that:

1. GAd DOES re-evaluate a site periodically, POSSIBLY based on CTR, content, as well as advertiser feedback and adjusts the EPC based upon it. This, at least on our site, appears to happen each Saturday morning. The EPC goes up or down and stays for the following week. This is NOT to be confused with changing ads and end-of-month budget cutoffs.

2. A CTR below 2.0% is detrimental to EPC, from 2-7ish is "ideal" positive and much above 7 is detrimental again.

3. Our EPC followed a pattern like you describe of getting better over time as if "more-trustworthy" but I feel there are a LOT of other factors that could cause this effect, such as getting better at optimization and zeroing-in on ideal keywords.

4. GAd obvisouly applies a zeroing-in algorithm which tests new keywords constantly and uses clicks (CPM) as feedback to determine the ideal ones. In this algo they always optimize for CPM first, followed by CTR (as opposed to EPC). Therefore EPC is the least important to them (possibly why it is not even listed on their reports much to our chagrin). Theoretically the longer it runs the more accurate it gets. Changing the content of a page wrecks havoc with it, for better or worse (removing low paying keywords from a page can free it to roam looking for higher paying ones). Initial ads on new pages take into account overall theme of site once established.

5. The first 30-60 days can be highly fluctuating, so don't expect to identify any meaningful trends before the end of that period.

[linear]

That's very interesting. The idea of a breakpoint at CTR=2.0% is pretty consistent with my observations. I'll just say that if I spot CTR over 2.0% I'm usually smiling.