| This 369 message thread spans 13 pages: 369 (  2 3 4 5 6 7 8 9 ... 13 ) > > || |
|Smart pricing analysis |
Goodness, is it the 17th already?
Okay, Here goes...
I came up with a theory that smart pricing was not really all that smart and tried to figure out what the bot does to determine your "smart pricing"
I think that there is an algo that ties in your poor performing pages and uses them to weight your site for worthiness.
The more pages you track the lower your quality (to the bot).
On the 2nd I deleted all custom channels and saw a 400% jump in income. Went through the 4th of July weekend with better than average earnings. Daily income going up by 2 to 8 dollars a day.
Ctr and ecpm are doing better but slowly, with ecpm rocking down and up.
Payout per click went up from 3 to 9 cents and had 5 .19 cent clicks on search...unheard of in the past. (my average payout per click was 8 cents for my niche)
Noticed better ads began showing up that had not been there before and no crap ads that I could find.
Things slowed down and started rocking so I thought, "hummm, mr bot has found another way". So I started going though my site page by page and changing out the channel ads for regular ones and adding them to pages that did not have them...more page views...:), and have seen the ctr and epcm renew their upward creep.
This weekend was higher than the weekends I've been having for several months now.
For me it is working, I still have over 800 pages left to examine and check for adsense channel ads, (doing this in alphabetical order.)
I am not saying anyone else should do this, that is your choice. All I am saying is it is working for me and judging by the payout and the better ads I can tell the worth of my sites have gone up.
It's about freakin' time! :) Now, let me study this data!
So you're saying there is a relation between how the bot views your page with regular ads, as opposed to ads with channels?
You alluded to this once before...
but I wasn't clear as to how it might have been working.
With this latest presentation of your test results... I may just be moved to try this as well.
BTW, I can't believe Hobbs is late! ;-)
And oh yeah... many thanks!
[edited by: Chapman at 5:28 pm (utc) on July 17, 2006]
I feel that the lower performing pages one is tracking weighs heavily against the worth (or quality) of the site itself and will trigger lower paying advertisers (crap ads) to your pages.
Thanks for the reminder.:)
That must have been hanging in the back of my mind when I sat down and did a quiet think through of possibilities.
Another thing I would be leary of is all of the toys Google tries to get us to use.
I do not say they are being underhanded or trying to suppress our sites, I do not believe in conspiriracy theories.
I am saying that it is possible they are relying more on flawed algos than they should and Not using the toys Just MAY force the bot to do things the regular way...Here's hoping anyway!
If things start to slide again then the toolbar will go next.:)
Thanks for reveaing all :)
I'm not clear one something though. Are you saying that on channels that are poor performers you still have a banner, but don't put a channel on it? Or are you saying that you are removing the ad and the channel for low performers?
Regarding the smart price algo - as it's done by a guy waving wet seaweed around in the Google car park and then thinking of a random number, what do we really expect? It strikes me as a really whacky situation where they can put crappy ads on in place of real ads that pay, and then hit you with smart pricing for the privelege! Why on earth don't they tie up the algos so that if an ad is causing you to be smart priced then they stop showing it. I mean - how expensive is it to buy a second bit of seaweed and pay another employee to wave it around?
[edited by: david_uk at 5:59 pm (utc) on July 17, 2006]
I am deleting the custom channel the page is on AND changing the ad to a no channel. (not tied to a channel, just a regular ad)
I took it to mean just removing the channel information so Google isn't alerted to possible low performance while maintaining the ad AND it's revenue. That's the beauty!
Is this correct Ann?
Sorry Ann... you ARE quick!
Okay maybe I can make it plainer.
#1. Delete all custom channels
#2. Go through site and find ads that was on those channels and change them to a regular-non channel- ad.
I am considering deleting all url channels too but I really think they are necessay to seperate the sites for my own knowledge.
It is a very simple plan--sometimes those are the ones that work :)
Thanks Ann, this is a very interesting theory. Do you think you have to delete the custom channels altogether or would simply removing them from the page work? Surely removing them from the page would be enough to hide the information from Google. If I decide to maybe try this it could be a sticky issue to delete custom channels as I'd like to keep them for past analysis.
Thanks again. Are you sure you are not a disgruntled ex-member of the AdSense team :)
Someone explain to me WHY Google needs to look at the channels you set up? Why would a publisher's choice of channels enter into their decision-making process in any way?
|Why would a publisher's choice of channels enter into their decision-making process in any way? |
This puzzles me as well. It doesn't make any sense to me.
It seems to me that Google should have the urls that adsense is called from anyway.
While I had never thought about it before, when trying to determine WHAT Google was using to evaluate performance, this certainly is a metric that they have at their disposal!
Perhaps it is not used for a Smart Pricing evaluation, however, since they are already providing this performance information to the publisher... why wouldn't they make use of it themselves? Pretty convienent I'd say!
As an additional thought... didn't Smart Pricing fall on the heels of an increase in the number of custom channels? Maybe Ann has a good point about being wary of each new tool we're given.
[edited by: Chapman at 6:48 pm (utc) on July 17, 2006]
I really don't understand that mechanism that's supposed to be at work here. If Google is using the channels we created as a factor in smart pricing, wouldn't that HELP some pages (the above-average ones in terms of conversion/perfomance) at the same time as it HURTS others (the below-average ones)? Or is your theory that they are simply penalizing for performance below a certain level, if found in a channel or channels, and applying this penalty to the site as a whole?
Not having channels designated sitewide should just cause all the pages to be averaged together, instead of being smartpriced separately, shouldn't it?
Am I missing something?
[edited by: hunderdown at 7:05 pm (utc) on July 17, 2006]
It all depends on how they might use the information. I never consideredt Smart Pricing to be an up<-nominal->down scenario only nominal->down hence no HELP... only HURT. They certainly could use it to HELP if they so desired.
All this, of course, a theoretical exercise!
Thanks for sharing Ann.
Now, not to sound like a dunce, I just want to make sure I understand your theory and test.
Does this mean that if I want to test your theory out on my own sites, I should go and delete all channels? As in, delete the channels by removing the adsense code from my pages and replacing it with code that does not use a channel?
Also, when you add any future ads, you would just not use a channel at all?
I just want to make sure I am understanding this. I do know as well, that what works for one site does not always work for another site. Also, as I've posted several times, I have something odd that happens, is that sometimes just changing the title color, even if slightly, sometimes the eCPM gets higher. It's very weird, makes no sense, but it happens, so even if your theory does not make sense to some people here, it may just work, even though it's hard to understand why.
I guess as always, we really have no way of truly understanding why our CTR and eCPM and earnings can jump around so wildly. I have for the most part the same ads which are very relevant to my site. From one day to the next I never know what to expect.
I hope this is something that works out well for you Ann and for anyone else who gives it a try.
I'm going to try it out and see if it works for me. Please let me know if I am misunderstanding how to undergo this test.
OK, and to add a specific question to Ann:
|The more pages you track the lower your quality (to the bot). |
And what if you only tracked the highest-performing pages on your site? Would that improve your smart pricing? Seems like a system that could be "gamed."
The bottom line is that poorly performing ads will dilute your smart pricing... No arguments there. However, ad channels HELP you identify those weak ads, not help G use it as a metric against you. If you truly believe that their efforts circle around ad channel metrics, I suggest you go and play with AdBrite. G can calculate by the url, the directory structure, the last time you took a crap, and by every other angle unbeknown to us, why would they resort to something so rudementary as the ad channels.
Listen, you might hit a quick crest by removing the ad channels, however, when G recalibrates your smart pricing, you are gonna be in trouble because you will not know which ads are weak and which are strong. Then you will have someone on here to thank...
G is not stupid. Yes, they are big brother, yes they are authority, and yes we have to kiss G's babies, but adsense is not a toy. It is designed to build up penny by penny, and when you drop the ball you deflate rather quickly... The first goal is to dispose of ANY poorly performing ads. Second, look at the ads showing by contextual versus site. Third, focus on improving the CTR:ECPM by ad units, not by page. The higher the CTR to ECPM, the higher your ECP will be. It wont happen over night, but thats a proven formula. The rest is you bringing traffic in and getting your visitors to click... It's a slow walk in the park, but G wont let it happen otherwise. And thats what smart pricing is about man...
[edited by: Yippee at 7:25 pm (utc) on July 17, 2006]
|Seems like a system that could be "gamed." |
Only if you were AWARE that channels were the cruix of the analysis! If they ARE then the system might easily be "gamed."
|why would they resort to something so rudementary as the ad channels |
Because that data is at hand! They've done the tracking, storage and analysis.
Why not use the data for their own (and their advertisers) best interests as well?
First things first, an apology is due here, also an explanation that I was only on your case out of impatience.
Second, thanks, and yes it is very simple as you said.
Third: Jomaxx & Jean: Yes Google has access to each URL performance, but Google has no way but channel definitions to tie up pages by topic or in relation to one another without our channel definition, if they want to evaluate our pages by url only fine, but to price or rate the performance per topic or section then your channel definition is the logical way to go.
Forth: Ann, while it sounds like a voodoo herbal remedy method, I like it because smartpricing as David pointed is witchcraft anyway and it obviously worked for you, I remember an old discussion about deleting unused channels, I will do that as well as your tips in a limited section the next time my earnings are down.
Finally: Did I thank you? If not then thanxx
BTW, while this is a very interesting theory, it would (should) be quite a "leap-of-faith" to just arbitrarily delete all channel information... especially for those who have been collecting data for years.
There is the possibility that simply breaking the channel/page connection may be enough for this to work as from that point the channel data for the new day will stay at zero.
I'll bet there are ALOT of zero channels out there because they've never been used or they've become unused. Google probably ignores channels with zero data and no associated active ad... I would!
Just a caution before doing anything too rash-
[edited by: Chapman at 7:53 pm (utc) on July 17, 2006]
You guys are taking this with a grain of salt... I personally think it will set you back months if you are dealing with big numbers. ANYWAY, I would definitely pay to see someone else take the so called "leap of faith" lol...
Let us know how it goes Chapman! You got guts man.
[edited by: Yippee at 7:36 pm (utc) on July 17, 2006]
No guts needed... simply postulating on Ann's discoveries on a 100 degree July afternoon.
No "leaps" until I've thought on it a bit longer but I'll likely implement some form of this no later than tomorrow.
[edited by: Chapman at 7:57 pm (utc) on July 17, 2006]
I would definitely try it, but I don't have a high enough CTR to notice huge changes.
I think Ann's theory has some merit though.
I don't think we should knock the idea. I don't know if it works, I don't know if it can be repeated. I guess at some point others will try it, and hopefully report back.
I don't have an opinion on this idea, but I do have an opinion on the knocking of ideas. In the dim and distant past I've posted results of the various things I've found that have worked for me in this forum. I've had a few scathing comments for my ideas (don't worry - I'm a big boy and don't get offended easily :) ). However, as it happens despite a load of doubt I was right. The two subjects I was head of the game on are removing poor performing ad blocks and zapping MFA's to increase income. These are now accepted as standard techniques to increase earnings. Maybe this one will work, maybe it won't but I do think we should discuss it objectively.
Oh - and likening smart pricing to witchcraft:- Nice comparison, but I think that witchcraft is far in advance of Random Pricing(tm) :) BTW, the witches are my favourite Discworld characters.
I said at the beginning of this thread that It was my theory and others could try it if they would like. There was no reason to become insulting in your post.
I believe Chapman, I think it was, maybe on to something by simply breaking the connection at the page level, changing the ad, but the old data would still be there. If you truly want to try it then do what I did. I printed out all my data for a year before changing anything.
I wish I was more knowledgeable concerning algos and bots, etc but I am not so I had to rely on logic...and before I get hollered at about that let me say that I did logically think this through before doing it.
Old unused channels that was hanging empty was deleted awhile back and I say a little upward movement then.
There are many ways to do things but NOONE has been able to report on something that actually worked for longer than a week. This is two weeks and holding upward today.
I am suffering from a bad case of eye strain. Can't hang in here too long. Please post how yours turn out should you choose to do it.
Sorry David_UK did not see your post..yes your ideas were good ones :)
Hunderdown, who says some of those adsense Millionaires arn't already gaming the system?
[edited by: ann at 8:21 pm (utc) on July 17, 2006]
| This 369 message thread spans 13 pages: 369 (  2 3 4 5 6 7 8 9 ... 13 ) > > |