homepage Welcome to WebmasterWorld Guest from 174.129.103.100
register, free tools, login, search, pro membership, help, library, announcements, recent posts, open posts,
Become a Pro Member

Visit PubCon.com
Home / Forums Index / Google / Google AdSense
Forum Library, Charter, Moderators: incrediBILL & jatar k & martinibuster

Google AdSense Forum

    
Innaccurate performance reports
Some adwords performance repors are wrong
graeme_p




msg:4351507
 5:39 am on Aug 15, 2011 (gmt 0)

I have two add units that are shown on the same pages. They have wildly different numbers for page impressions, and the one that is below the fold is shown has having much higher CTR and RPM (eight times the RPM!) in the ad units and ad sizes reports.

These numbers are also different from the more believable numbers I get from the custom channels report, so one or the other is definitely badly wrong.

I also find it hard to believe that I am getting five times the RPM from CPM bids that I do from CPC bids.

Can we rely on these numbers? More specifically, is it worth trying to optimise on the basis of these numbers when they are obviously sometimes wrong.

 

alika




msg:4351585
 12:27 pm on Aug 15, 2011 (gmt 0)

getting five times the RPM from CPM bids that I do from CPC bids


Many here are saying that they want to block CPM ads and run only CPC ads. Or that they only want text ads, not image ads

But my numbers are saying that

- CTR of CPM bids are nearly 2.5x more than CPC ads
- RPM of CPM ads are 8x more than CPC ads

If anything, I want MORE of CPM ads than CPC ads

Same with image and text ads

- CTR of image ads is 2x as much as text ads
- CPC of image ads is 2x as much as text ads
- RPM of image ads is 5x as much as text ads

I want more image ads!

Play_Bach




msg:4351608
 2:36 pm on Aug 15, 2011 (gmt 0)

> I want more image ads!

Alright alika, I've just enabled image ads again. They haven't worked well in the past, but always willing to do a test.

netmeg




msg:4351612
 2:48 pm on Aug 15, 2011 (gmt 0)

Yea my CPM ads are way up this year. I have to think it's because of all those other networks.

graeme_p




msg:4351726
 6:23 pm on Aug 15, 2011 (gmt 0)

But are these numbers accurate? I know that at least some are wrong, so why should I trust the others? Maybe my (and alika's) CPM ads are not high earnings and its just an error.

Sally Stitts




msg:4351742
 6:49 pm on Aug 15, 2011 (gmt 0)

What is RPM, besides Revolutions Per Minute?
I've never heard of this metric before.
Is it Revenue Per Thousands of Impressions?
[acronymfinder.com...]

Is it similar to eCPM - Cost Per Thousand Impressions?
What is the formula?
Please clarify.

It is difficult to buy into your statements, when you use an undefined metric.

EDIT: SORRY. I just checked the NEW system. I don't use the NEW interface, where all is made clear (almost).
The NEW INTERFACE uses -
RPM = (Estimated earnings / Number of page views) * 1000
whereas
The OLD INTERFACE uses -
eCPM = (Earnings / Impressions) * 1000

I wonder why they changed this. Can anyone guess?
For one thing, page views and impressions are not always equivalent. Some page views don't get impressions.
Also, how are "Estimated Earnings" and "Earnings" different, as far as the formulas are concerned?
Thanks.
.

[edited by: Sally_Stitts at 7:16 pm (utc) on Aug 15, 2011]

alika




msg:4351744
 6:57 pm on Aug 15, 2011 (gmt 0)

Sally Stitts .. if you actually use the new reporting system, the eCPM has now changed to RPM. It is not an undefined metric

Here's Google's definition of RPM [google.com...]

Sally Stitts




msg:4351748
 7:15 pm on Aug 15, 2011 (gmt 0)

No, I don't use the new reporting system.
Which explains my ignorance, as you suggest.

bumpski




msg:4352660
 1:26 pm on Aug 17, 2011 (gmt 0)

graeme_p

In many cases you must use the "columns" drop down box in the upper right corner and choose "ad requests" or "individual ad impressions" to get accurate impression statistics. The PageViews mode tends to attribute all the page view data to the first ad unit in the report. Frankly it seems to me they could correct this.

Regarding RPM and eCPM

I think Google introduced RPM to eliminate confusion between terms used for Adwords versus Adsense. For CPC ads Google must calculate (guess) an "Effective CPM" value so they can compare bids between CPM and CPC ads. (How accurately they do this I sometimes really wonder, especially for low traffic pages!) The Adsense RPM (formerly eCPM column) is much simpler than this, and is just Earnings per 1000 Impressions. Below a link to Adwords help regarding "eCPM" which really is nothing like RPM.
[adwords.google.com...]

graeme_p




msg:4352866
 10:48 pm on Aug 17, 2011 (gmt 0)

Thanks bumpski, that helped me A LOT.

I now get COMPLETELY different results: CPM ads pay less, image ads pay much less, the ad below the fold pays much less per ad request. Its slightly different per impression, which is interesting.

bumpski




msg:4352876
 11:30 pm on Aug 17, 2011 (gmt 0)

From my Adwords help link:
No matter which type of ad wins the position, the AdWords discounter monitors the competition and ensures that the winning ad is charged only what is necessary to maintain its ranking above the next-highest ad.
What's really upsetting about this?
Imagine being at an public auction, you bid $1.00; your opponent bids $2.00 to "shut you out", and wins the auction.
Then you learn that the winner only had to pay $1.01; one cent more than your bid!: and that's the way Google's Adwords auction works! Great for the advertiser!
Who ( I believe ) in 2003 actually had to pay what they bid! (And what's worse, they were happy to do it!)

Google threw the earnings away......

netmeg




msg:4352940
 2:59 am on Aug 18, 2011 (gmt 0)

Advertisers specify a MAXIMUM bid; doesn't have to actually be what they pay. I routinely specify a maximum bid that's hight but actually pay quite low. And this is how you attract and keep advertisers in the program, without which you would have no AdSense. I for one would not advertise at all if it were otherwise.

graeme_p




msg:4352947
 3:59 am on Aug 18, 2011 (gmt 0)

@bumpski, the idea is that you do not have to worry that you are bidding higher than you need to. Suppose you are at an auction, you cannot see the other bids and are willing to pay $2. You estimate a 90c bid will win, so you bid 95c. You then lose because of a $1 bid. This way, you can bid $2, and know that if 90c would have won, that is what you pay.

Its necessary for adsense given you cannot see the other bids.

graeme_p




msg:4353345
 2:57 am on Aug 19, 2011 (gmt 0)

Individual ad impressions numbers are wrong as well: an ad that appears no more than once per page can receive more impressions than the number of page impressions.

Play_Bach




msg:4353413
 10:45 am on Aug 19, 2011 (gmt 0)

None of my stats are in agreement. Click count, impressions, earnings are all over the place as I go from report to report.

bumpski




msg:4353809
 5:05 pm on Aug 20, 2011 (gmt 0)

I think some minor quirks are due to the use of "back" browser button. I think historically Google did not count a re-rendering of the page caused by the "back" button as a "Page View", but it is likely they may show different ads as a site visitor "backs" out. Same PageView new ads ?

There's got to be lots of minor quirks from pages re-rendered from cache.

Regarding inconsistencies, going back to 2003,2004, etc. Google would introduce massive errors in statistics, especially PageViews off by a factor of two or three X. When I compare to old, old stats I've got to remember all these huge quirks!

Global Options:
 top home search open messages active posts  
 

Home / Forums Index / Google / Google AdSense
rss feed

All trademarks and copyrights held by respective owners. Member comments are owned by the poster.
Home ¦ Free Tools ¦ Terms of Service ¦ Privacy Policy ¦ Report Problem ¦ About ¦ Library ¦ Newsletter
WebmasterWorld is a Developer Shed Community owned by Jim Boykin.
© Webmaster World 1996-2014 all rights reserved