Forum Moderators: martinibuster
Why isn't the earnings more important?
I'm running a webshop, started up late november 2005
I put adsense on my site two days ago.
CTR = 6,5%
eCPM 29,01
Is this good numbers, bad numbers do I want my eCPM to be a big number or a small number?
As for the eCPM that is what your rate of pay per thousand impressions are. Obviously the higher this is the better. Would you rather have a site paying you 30 dollars a thousand impressions or 300 dollars a thousand impressions?
Regards
ricey
Best of Luck and Welcome to the Forum
Statements about the importance of eCPM are usually in response to publishers who obsess about their earnings per click, or EPC. eCPM is a more meaningful number than the number of cents or dollars per click from a publisher's point of view (as opposed to an advertiser's point of view). Why? Because eCPM can be calculated for any revenue stream, whether it's AdSense, banner ads, affiliate sales, direct ads, etc. This lets you compare different revenue sources and see if you could be using your "screen real estate" more effectively.
Also, if you set up channels for different areas of your site, you can use eCPM to see if AdSense is working for some topics or content types but not for others. (Hypothetical example: You might find that AdSense has an eCPM of $5 on product-review pages but only 50 cents on forum pages. If that were the case, you might be better off replacing AdSense with something else--such as CPM banner ads--on the forum pages.)
Maybe I'm a little obtuse but I honestly don't get this whole eCPM business
EFV explained it pretty clearly in the post above yours :)
If you want to compare ad companies and earnings (since not all ad companies pay per click), the eCPM will give you a clear ruler to measure by.
If you just use adsense and aren't diversifying your income on the site, then eCPM might not be as important for you.
If you just use adsense and aren't diversifying your income on the site, then eCPM might not be as important for you.
It's a pretty handy metric even for internal use, especially if you're using channels to track different pages or areas of your site. Example: If you've got a travel site and you find that the eCPM for AdSense ads on your photo galleries is abysmal (compared to the eCPM on your text pages), you might decide to use the AdSense space for another purpose, such as links to the areas of your site that produce better revenue.
Let's use a real world example. From what others have posted, YPN is often paying more per click. However, because their ads aren't usually as well-targeted as Google you are likely to get lower clickthrough rates.
If you get a 10% clickthrough rate at $0.10 a click it's the same as a 1% clickthrough rate at $1.00 a click.
In the end, eCPM is the product of clickthrough rate multiplied by the average click value multiplied by a thousand. It's why it's more effective because it takes it all into consideration.
publishers should have full control over the minimum payment per click that can be displayed on their sites, and there should be full disclosure on which advertiser is paying what, per click.
It can be amusing to calculate your EPC and think that you're getting 50c, or some other high number, for every click. But if you could run 20c ads that are better targeted and more compelling, so that 4X as many people click, wouldn't that be better? Of course it would.
But yeah, other than that I haven't found much use for it. :-)
JK
eCPM is not hypothetical, it's what you are actually earning. For measuring your AdSense performance it's actually a better measure then your bottom line earnings, because that number is obviously dependant upon the total traffic to your site on a given day.
Bingo! That's why, if my traffic drops on a Saturday (as it always does), I don't panic over the fact that my earnings are down and post a message about it on Webmaster World. :-) I can see from the eCPM that AdSense is performing normally.
The eCPM figure will show that the ads on the page in question perform really badly in comparision to other pages. A real world use for the figure is in the decision whether it's worth running ads on that page or not. You may get overall better income from your site by dumping the ads on that page and increasing the overall site ctr.
A real world use for the figure is in the decision whether it's worth running ads on that page or not. You may get overall better income from your site by dumping the ads on that page and increasing the overall site ctr.
I'm not so sure about that (the idea that dumping low-eCPM pages will improve total AdSense income), but knowing the eCPM of pages or a site section can make it easier to decide whether you might be able to use those pages more productively for something else.
Even if you don't use that page for monetisation in other ways, not having ads on every single page may help decrease ad blindness for the pages where ads do work well.
Thanks for the input everyone.
eCPM is not hypothetical, it's what you are actually earning
This statement is converse to the truth.
Although eCPM is hypothetical, it is a very useful figure in some areas, eg in comparing revenue streams. It is not so useful in other areas, eg: when assessing the impact of changes that you make to your site.
Imho, all the numbers we are discussing are useful sometimes - not only eCPM and Earnings, but also Average Earnings Per Click, Click Through Rate, etc..
However, one has to choose the right metric to support the particular analysis or decision being made. So, to respond to the OP's question, I would say that eCPM is a useful number, but it is not "the most important thing".
If smart pricing takes overall CTR into account (and logically it should use it as a factor)
One can't know for sure, but the statistical analysis I've done suggests that CTR is not taken into account in smart pricing: over a long period of time, on our sites, the relationship between EPC and CTR is a purely random one.
It may work differently on individual sites, of course, but I would recommend doing some hard statistical analysis to decide, including testing for "significance" of the results. If there is no 'better alternative' (and we don't have YPN in the UK yet) then there is the potential to lose money. Our 'poor performing channels' collectively add up to a lot. It's like cricket - 4s and 6s are nice to have, but the bulk of the score is usually made up of singles.
One can't know for sure, but the statistical analysis I've done suggests that CTR is not taken into account in smart pricing: over a long period of time, on our sites, the relationship between EPC and CTR is a purely random one.
It's all supposition of course. I've just looked back at my monthly data since Jan 2004. In 22 out of 24 months, when CTR went up over the previous month, EPC went down and vice versa. Hence it's my gut feeling that somewhere along the line it is taken into account by smart pricing, but of course we will never know for sure!
eCPM is not hypothetical, it's what you are actually earning
This statement is converse to the truth.
* What you are "actually earning", in Adsense Reports, is in the column called "earnings"
* eCPM is a calculated figure, the method of which is explained in the Adsense help pages, and is therefore 'hypothetical'.
What you are "actually earning", in Adsense Reports, is in the column called "earnings"
LOL. It's amazing how many people miss that. :-)
I actually have several channels like that. One channel gets maybe 1 impression a day, so for the last 2+ years, eCPM has been hypothetical in that sense.
Maybe I'm wrong, but I think this is where the "hypothetical" comes from.
JK
That's not what "hypothetical" means. The speed of a car is a calculated figure based on time and distance travelled, but there's nothing hypothetical about that either.
Well, although the speed of a car is subject to Einstein's theories and the choice of units (miles and kilometres), compared to eCPM it is relatively certain. The eCPM calculation is based on some frail and changeable assumptions.
You can see the frailty of those assumptions, for example, by changing the "show data by" option in your advanced reports. Show "by page" and you get one number for eCPM, show "by individual ad unit" and you get another (unless you have one ad unit per page, in which case they will be the same).
Another way to illustrate the frailty of eCPM assumptions, or the disconnect between eCPM and actual earnings, is by removing low eCPM channels/pages from your advanced reports: the eCPM figure will go higher but the earnings figure will go lower.
A third way, but the feedback is not so instant, is to put Adsense on, or remove it from, pages where you get no clicks. You will see the site eCPM figure change, but earnings will remain the same.
The biggest problem with eCPM is that it is based on the hypothesis that the number of impressions used in the calculation corresponds to the volume of traffic at your site. This hypothesis can be wrong for a number of reasons, such as site speed or stickiness.
eCPM is not quite "how long is a piece of string", but neither is it as certain and obvious as the speed of a car.
Are you just arguing for the sake of arguing? The points you're making are irrelevant, even laughable.
I don't see how saying "actual earnings = earnings, and not eCPM" is laughable. If anything the converse would be true. I am making a serious and valid point, my intention being to help you and other members of the forum make more money through more informed decisions.
However, as the dialogue seems to have moved away from rational argument, I'll leave it there.
I actually have several channels like that. One channel gets maybe 1 impression a day, so for the last 2+ years, eCPM has been hypothetical in that sense.
Maybe I'm wrong, but I think this is where the "hypothetical" comes from. >>
This is exactly what I meant when I first mentioned 'hypothetical'.
In addition to allowing publishers to compare relative earnings from different advertising programs, you can also use eCPM to perform your own optimization testing to ensure you're making the most from your AdSense account. Total earnings are not always as meaningful in this respect.
For example, say you're experimenting with ad formats (or colors, location on the page, etc.). For 2 weeks you use a Rectangle format on your homepage and use channels to track the eCPM. For the next 2 weeks, you use a Leaderboard format and use channels to track the eCPM.
In this way, regardless of changes in traffic, etc., you can see which ad format actually performs better on your site and apply that lesson throughout your site. eCPM can be a very useful for performing scientific experiments to maximize your site's revenue potential.
