Forum Moderators: martinibuster
I had 10 different channels and now created a new one called "all-sites". I had to update the code on all of my sites and pages, whew... took a couple of hours to do it.
I left the old channels intact, in case this doesn't work out, I can go back to the way it was.
Will post the (hopefully positive) results in a week or so.
Folks, this is not worth discussing. We will unveil something true here. If Google finds out that we know already, then they will change the way they monitor our websites.
Just keep your discovery secret. Cheers!
I couldn't believe few are participating this thread. I assume they discovered this already and they don't want Google to know about this that publishers have unlocked their channels secret.
What secret? Some guy messes around with his channels and suddenly has an earnings spike. 2 days later his earnings have dropped to an all time low and he's quitting adsense. If that's the secret I believe I'll pass...
However, I do understand it from Google's perspective. If you want to remove low performing ads, you channel things. Helps you mess with their algo. Which they don't like. It also potentially helps you test the revenue share. With possibly fraudulent clicks, something else they don't like...
The end result, in my mind, is that you can't trust channels (not talking url channels).
Changing the payout on this basis is a very bad thing I think. What if you have a partnership with several people and use channels to track payout rates? Setting up a channel should NOT affect the income.
I think it's entirely possible that Google uses channels to determine smart pricing, but that would only benefit the publishers, not hurt them. Instead of smart pricing the entire account, they could reward the specific subsection of ads which are truly effective.
Also, I think that it's far too common for people to use anecdotal data to try to make conclusions about how AdSense works. For those who are serious about trying to decode what AdSense is doing, it's going to require a large number of people working together, sharing and aggregating and crunching data, which is probably against AdSense TOS anyway.
Amazingly enough, I didn't try this experiment over the weekend and my ECPM rose by over 30%.
But I also didn't experiment with it the 3.5 weeks beforehand and during that time my CTR fell by 20%.
But then when I think of Christmas, strangely another time that I wasn't trying it, my earnings were double what they were...
Dear diary, my head hurts.
However, now with the .27 ecpm, i get a daily earning of around 180, instead of the usual 130. So i have a gain of about 50. But still very far from last year's 300.
Im going to experiment with channels again. Remove my remaining low performing channels altogether and only leave channels that have .40 ecpm or above rate.
The weirdest part is I spent so much money from advertising since the last 5 months when i notice a steep decline of my adsense earnings. I thought thru advertising my site, I will get page views and more adsense revenue. I was wrong. My page views has increased , almost doubled what i had last year but my earnings just keeps going down.
I also have 2 other 3rd party ads on the site so it could also play a role in my adsense revenue going down. I always believe that i should not put all my eggs into one basket but I'm not sure if that is a good idea. I think my 3 party ads are stealing some of my adsense revenue. As a result, Adsense robot recognize that and reduce my ecpm. I wonder what would happen if i remove my 3rd party ads, will my ecpm go back to last year's figure of .68? Maybe. But then again, i don't want to lose my 3rd party ads for now because I always think of the worst case scenario. If Adsense ever drop me (though very unlikely), I still have my 3rd party ads to fall back to.
wonder what would happen if i remove my 3rd party ads,
The hawthorne effect found that ANY change in the environmnet changed the outputs.
It was found when they were adjusting the lighting in the Hawthorne factory and found no matter what they did output increased.
In this case, it wasn't long ago when everyone was clamoring over making new channels and dividing the site up. Threads as postive and assertive as this were saying how much their income increased.
Tell me about it a few weeks down the track, I am getting tired of jumping on these wagon shows.
We shouldn't be changing our sites around as if Google was some strange black box, (I don't know why it works but it does!) where if we just get mantra right on the input site we get the prize on the output side, like a cargo cult.
There must be logical, practical, verified methods for adjusting our sites that help all us struggling drowing webmasters, that we can institute instead of this clutching at driftwood as we drown under decreasing ecpms....
March: Exceeded $200/month for the first time.
April: Exceeded $200 for first half of april! (on 4/15)
Also exceeded $25/day this last week. (It's back down, but the lowest low is about double my old average from a month ago..)
The 'spring break' / Easter time frame has been incredible for me, and I can't figure it out, because my sites aren't really about youth-oriented topics.
Only real change: I cleared out a bunch of old entries from my competitive ad filter. May be a coincidence though.
Only real change: I cleared out a bunch of old entries from my competitive ad filter. May be a coincidence though.
How did you figure out that these were "old entries" (which I read as "entries that are not relevant for your site(s) any longer")?
This thread has gotten linked. I'm sure Google noticed it. They may have (temporarily or permanently) changed the algo.
The only data that is relevant are the early ones that were done prior to this thread getting picked up.
The principles I go by are that Google likes collecting data and spreading FUD so that you can't.
It's clear that channels have historically impacted earnings. I've personally seen it with every test I've done. They don't want you to know what improves earnings, period.
They don't want you to know what improves earnings, period.
They don't want you to know what improves earnings, period.
And how is that approach beneficial to Google - when I, as an AdSense publisher earn more, doesn't Google earn more?
I've experienced a downward spiral in earnings recently like a number of other people have reported here. I've also experienced recent lousy targeting on pages that have been getting good targeting for years.
I've tried different techniques, implemented suggestions I've found at Google, etc. It mostly seems like I'm throwing things at the wall to see what sticks for lack of solid information.
As a result, in the past few days I've decided I don't have time to waste on such a guessing game and some of my space that was formerly provided to AdSense is now provided to other sources of revenue.
Some of that space has been switched to selling my own products and I'm spending more time working on my own products.
So to repeat my original question, and it may be just a rhetorical question, is how does Google benefit from this lack of communication if people like me stop spinning in the mud and slowly start walking away?
FarmBoy
If you figure out Google's algo, you will tweak your site to make a bigger share of an existing Qty. Which does not translate into better numbers for the advertiser. It's basically taking a better share from another publisher.
That's my take. You can argue the point. And I agree with you. I think channels shouldn't be part of the algo. I'm just trying to understand it from their POV. Because I've noted that it's extremely hard to understand what's going on. To "get" the smart pricing algo. Obviously they are obfuscating it....
The best advice I've heard on this is to build sites that get good conversions. Review sites are best at this. But then again, who wants to change the kind of site they build just to make Google happy? Why not become a baseball player, they make a lot of money...but maybe you don't want to be a baseball player, or aren't good enough to make it, same way I'm not interested in a review site nor do I want to put in the effort to make it succeed.
And how is that approach beneficial to Google - when I, as an AdSense publisher earn more, doesn't Google earn more?
Let me put Clark's words in more simple terms:
Google earns the money anyway, the same ad on your site or someone else's, the money is already in their pocket.
But to say that "They don't want you to know what improves earnings" is a bit too harsh, they honestly want you to earn more, but not at their expense (lowering the quality and trust in their network), or at the advertiser's expense (click fraud), or at an unfair disadvantage to other publishers like if you call for attention to ads, or know a trick that hogs all the best paying ads to yourself forcing other publishers in your niche to seek competing networks..
What Google probably wants is for the whole network to grow proportionally with no spikes, they probably hate spikes and surprises, and I am willing to bet that they have a dampening element in their algo, I would if I ran AdSense.
Google earns the money anyway, the same ad on your site or someone else's, the money is already in their pocket.
I agree with this to a point, but not beyond, because I don't believe the size of today's advertising pie represents the maximum possible.
FarmBoy
it used to read:
google_ad_channel = "";
now it reads:
//2007-04-1X: NAME OF CHANNEL
google_ad_channel = "Some Random Number";
This change was done over the weekend. I guess someone read this thread. I think this means there was a good chance there was a bug or there is a new anti-gaming tweak to the algo.
In other words it was already like your send snippet hours before the update.
I have now added a link "Advertise on this site" on all my pages as suggested by Adsense blog. You can follow the instruction here: [google.com...]
In other words it was already like your second snippet hours before the update.
But certainly the change was made after this thread began because I had some of the first snippet on code I generated on the day the thread began.
How did you figure out that these were "old entries" (which I read as "entries that are not relevant for your site(s) any longer")?
Old as in: I'd set them up many months ago, when I had a few new sites, G hadn't yet had time to figure out what the new sites were about, and all I was getting was ringtone and other completely off-topic ads. My experience, over time, is that once the site's topic is known to G, most of the off-topic ads seem to disappear from my sites without the help of the competitive ad filter.
Of course, I still need to monitor the situation, but I've not seen any of the old offenders since I cleaned out (most) of the filter three weeks ago in response to G's 'April Optimization Report' suggesting that my filter might be costing me money. Most of the filter entries were not MFA sites, as much as just strange and off-topic sites that made my site look lame.
I really think the filter thing is a red herring (at least for me), because the advertisers I removed from my filter aren't showing on my sites as far as I can tell, so how can clearing the filter increase my revenues in this particular case? (Yes, I know I can't know all the ads that are shown to other users...)
YMMV.
Hmm. It seems as though the channel code was changed
this change is weeks not days old, I've been seeing it for a long time.
Hmm. It seems as though the channel code was changed
I've been seeing it for the last month. It predates this thread by at least 3 weeks and probably longer than that.
You guys need to let this one go and get to work on that Kennedy thing...
