Forum Moderators: martinibuster
Publishers frequently come on the board lamenting that their earnings have decreased markedly since a certain date. Might Google periodically audit publishers to determine how well the traffic they send through AdSense is converting? This would be the reason why, a few months back, Google started encouraging AdWords advertisers to use their conversion-tracking tool.
If a publisher's traffic is not getting the job done they could cut the revenue share they give to that publisher (and cut it more later until it is comparable with the quality of traffic they are producing or the less desirable publisher takes his traffic to another contextual advertising network). Some of this extra money would presumably go in Google's pocket and some would probably serve to increase the percentage paid to more desirable publishers and entice them to send more traffic.
This would be a methodical way of retaining the publishers they deem the most valuable and weeding out the undesirable traffic producers without having to completely terminate their relationship.
Sorry if this is common knowledge to everyone on the board but I don't think I fully grasped the tools AdSense has at its disposal to control the quality of its publisher base until now. I always assumed all publishers got the same percentage. Could we be seeing the beginning of different "classes" of publishers?
quite frankly, conversion rates have more to it then quality of traffic, a lot more, eg:
1, site could have usability problems.
2, site could be very expensive
3, site could be used as an information centre.
4, most adwords advertisers do NOT use google conversion tool.
there are so many reasons that to me are obvious, that this is NOT the reason for lower payout levels.
Shak
Personally I think it would be immensely hard to do this on a scale large enough to be worthwhile.
<edit>They do offer some kind of conversion tracking tool, but from what I understand the information currently available to Google would be pretty crude.</edit>
If they have been around a while they will have some idea of what they're conversion rate is. Maybe it is 0% since they are an information site. If that's what they are happy with then fine. Let the campaigns run the same everywhere. If they have usability problems or are expensive then they would see a lower conversion rate from any advertising they do and might be happy with a 1% so fine let it run. Maybe they get a 50% conversion but on one site they get much lower so have the system lower the cost to them for that site only.
As long as a conversion tool is used and the advertiser has the option to say what they want from the campaign it works quite well. As jomaxx said though it is not an easy thing to create and I don't think Google is at that level yet with AdSense but someday they might. This is the way it should be done for the same reason everyone will have a different rate card for their site. Sites are just not created equal and should not be treated equal.
JAG
cut it more later until it is comparable with the quality of traffic they are producing or the less desirable publisher takes his traffic to another contextual advertising network
Making changes towards the traffic type might be reasonable, but the latter suggestion doesn't seem right as Google has not shown any hesitation in the past to remove Publishers from it's network when they had reason to remove them. So I don't think it fits in for them to suddenly beat around the bush to remove publishers with that method.
I was under the impression you can't do either with AdWords, both of which you would need to do to be able prevent your AdWords ads from being displayed on "less desirable sites" as the original poster hypothesized.
The only other option is that Google would be doing it automatically for all AdWords advertisers, but this would be a massive job.
thx
This wouldn't make any sense at all.
If the ads served on a certain publishers site don't convert well, then Google should serve them better matching ads, not cut their pay. There's really nothing a publisher can do about conversion rates.
But if the ads are targeted and they do convert well for another site that has similar content then why assume the ad needs to be better targeted?
There have been plenty of times when we set up campaigns at the marketing company I worked for and while the content and site topics were similar there were always differences in conversion rates. So, if we targeted something such as a financial channel on some sites it would work and others it would not even though they were close in content. It would not make sense to target sports to the financial sites that did bad just to see what would happen since the targeting was already what should have worked. It is other factors such as demographics and such that ultimately determine if an ad will work and AdSense knows nothing about that.
So the bottom line was to negotiate the rates down and run the ads so it worked for the advertiser or drop the sites that could never work. Sometimes it seems we forget it is the advertiser who pays Google and if they aren't happy then what?
JAG
But there have been several instances since the launch of AdSense that many publishers have posted here about drops in EPC centered around a specific date. I submit that the only way for that to happen (en masse) is for Google to change the payout structure.
Advertisers don't all drop their bids at the same time.
Publishers don't all change the content at the same time.
Advertisers don't all sign up the same time and flood the market with demand.
Publsihers don't all sign up at the same time and flood the market with supply.
Google is the only entity among the three that can make changes to affect sites across the spectrum.
I've also noted that these changes take place over a holiday or weekend, where payouts tend to dip anyway. Once the work week starts back up, the payout does not recover to previous levels.
The fact that they are not forthcoming about the payout structure would allow them to change it with little resistance from publishers.
Yes, I know when we signed up we knew we would be operating blind in terms of payout. But I think Google may not be considering the long-term implications. Dropping payout to low levels may encourage fraud, as sites try to recover income they perceive as lost. Publishers will feel helpless as their income drops with no explanation as to why (only speculation and commiseration here).
I have not seen AdsenseAdvisor's input on this issue, it would be nice if someone from Google would explain to publishers why EPC is dropping. Treating the relationship between publishers and Google can only be beneficial to both parties, and that includes being more forthcoming about payout. After all, revenue is the only reason publishers signed up in the first place.
But clearly Google is never going to negotiate down AdWords costs if an advertiser's conversion isn't high enough. It goes against the whole concept of bidding for top position, and would effectively turn them into a pay-per-performance network.
I'm not saying Google will or even can do it. I'm just saying that it is standard practice to do when you don't own the real estate an ad will be on. Been that way for years.
But they wouldn't have to negotiate anything down since under the current AdSense model the content site has no say in what they want for their advertising space anyway. They could pay $10.00 a click and charge the advertiser $50.00 a click and the content site would be happy as could be. On the flip side a content site that is used to getting $10.00 per click may now only get $0.10 per click because of available advertisers.
FindWhat has this capability built into their contextual advertising system and it works. Overture has a different pricing structure and it works. So it's just a matter of Google building the system to it's full potential which at the end of the day is not just pay the content site the maximum amount but to make sure the campaign works for the advertiser and then let the content site decide if it is worth it to run or go back to banners.
If it doesn't work for the advertiser then the program goes away as advertisers pull out then we're all back to banners whether we want to or not.
JAG
Advertisers don't all drop their bids at the same time.
I have to agree with this... One particular page had been showing ads that were targeted perfectly since creation {about a month ago}. I figured that the matching was so perfect that this one page could have been giving me double digit CTR. Of course, there is no way to find out :(
The thing is that on Sunday this page changed from perfect targeting to a similar yet not so related widget. The change was more from:
Steps to acquire golden fuzzy widget
to
Go to school or get a job as a widgeter <groan>
Nobody wants to be a widgeter, people want their widgets! Google give people their widgets!
What are the chances of 4 or possibly more advertisers changing their mind on the same day? The change here was very evident, but other changes might not be as much.
Regarding the main topic of the thread, I think that the drop in EPC has not dropped across the board. In fact, 4 out of 7 days showed higher than average EPC for me. I just don't want to remember the 3 bad days.
Dropping payout to low levels may encourage fraud, as sites try to recover income they perceive as lost.
Not really. A true automated system that can make adjustments in real time would see this a click fraud and continue to drop the payout until it eventually reached zero.
In other words, click all day long but if there is no return to the advertiser there is no payout. Perhaps that would even get rid of the dreaded email.
JAG
But there have been several instances since the launch of AdSense that many publishers have posted here about drops in EPC centered around a specific date. I submit that the only way for that to happen (en masse) is for Google to change the payout structure.
How many is "many"? A drop in the bucket, compared to the total number of AdSense publishers, and certainly not a statistically valid sample.
Anecdotal evidence is usually more anecdotal than it is evidence. :-)
lol. So true...
Well, I have mostly information and reviews sites that should convert fairly well, and EPC has been way down this last week. I'm hoping things will pick up with the first of the month.
My content is very heavy on one term.. one subject and the PPC has been consistently proportionate to the bids on Adwords.. in the recent days it's dropped about 30% while the advertisers have remained the same.. as well as the bids on Adwords.
I'm not sure why.. but maybe we'll hear from AdsenseAdvisor.
So even if there appeared to be a correlation between AdWords bids and your earnings per click, the ratio between the two might easily change over time whether or not Google made changes to its payout formula (depending, for example, on how many clicks you had in a given day, whether you'd reached a certain break point for the month, etc.--and that isn't even counting factors such as whether a major advertiser's inventory of clicks is being spread over a growing number of impressions on your sites and other sites).
IMHO, it's impossible to draw an accurate conclusion about Google's payout formula without more information and a large statistical sample over time. If your EPC is dropping--and, more important, if the all-important effective CPM and bottom-line revenues are dropping--it's more productive to find ways to improve those numbers than it is to try and reverse-engineer Google's payout algorithm.
It's true that there are various classes of websites, and thus various quality of traffic. It is also truth that generally payout from Google AdSense drops.
It's only "truth" if it's supported by evidence, and so far we've seen no evidence to support that claim--merely supposition by publishers who are looking for reasons why their revenues have fallen. (And revenues aren't the same as payout.)
Theres only one solution to these problems : increase your traffic, particularly in the sense of unique visitors (repetitive visitorsb are less worthy).
That's a good point, and sites that are very "sticky" (community sites, for example) may not be the best venues for AdSense.
Still, repetition doesn't always work against success with AdSense. A travel site, for example, might attract the typical user several times: once when he's looking for places to go, again when he's researching a specific destination in detail, yet again when he's looking for a hotel or a vacation rental or sightseeing ideas, etc. Each time, that user would be a good prospect for targeted AdSense ads.
Similarly, a site about amateur radio or digital photography might be "sticky" because it appeals to hobbyists, and those hobbyists might ignore ads most of the time--but when they're finally ready to buy new transceivers or digital cameras, they'll be hot prospects for AdSense advertisers.
Getting back to the payout issue, I think publishers should worry less about Google's payout formula (which is almost certainly more complicated than a simple percentage split) and focus on the bottom line. The real question sholdn't be: "What percentage am I getting?"; it should be: "Is AdSense a profitable way to use the advertising space on my Web pages?" Even if a publisher knew, for example, that another ad network paid 50-50 and Google paid only 65-35, he'd be foolish to change from AdSense to the other ad network if his effective CPM or bottom-line revenue dropped.
(FWIW, AdSense is currently generating at least 10 times the revenues that a banner-ad network used to generate on my site, so I'm better off with AdSense even if it has a lower payout...which may or may not be the case.)
IMHO, it's impossible to draw an accurate conclusion about Google's payout formula without more information and a large statistical sample over time. If your EPC is dropping--and, more important, if the all-important effective CPM and bottom-line revenues are dropping--it's more productive to find ways to improve those numbers than it is to try and reverse-engineer Google's payout algorithm.
I agree that concentrating on figuring out Google’s payout scheme… probably not as complicated as an algorithm, but a simple scaling… <although you can’t beat computer geeks at making things more complicated than they should be, and there is something to be said about integrating topics and themes into the whole scheme if you want to capture a particular segment> is not a productive thing to do. After all, the decision should be is Adsense still the best option for this adspace or can I make more money with something else?
Nevertheless, I think that if your website is fairly constant when it comes to EPC, and I don’t mean on a daily basis, but on a sizeable sample <ie: trailing 7 days>, and you notice a drop that is higher than anything you have seen in 7 months… Then you could make an educated guess that something has changed and the drop in percentage payout is as good as any other possible explanation.
</rambling>
It is also truth that generally payout from Google AdSense drops
I disagree.
I have seen the threads where several posters complain of a sudden drop in EPC on a certain day.
But JohnKelly, while I agree publishers are unlike to change content en masse, I don't agree with the other points particularly about advertisers not all dropping their bids at the same time.
This is not as unlikely as some may think. Several advertisers could pull out at the same time but ALSO if one or two big advertisers dropped their Adwords budgets you'd see the same effect
I have seen increasing returns from Adsense. There was a time EPC dropped but then it went back up and up. Perhaps at regular intervals Google moves the high paying ads from one "type" of content site to another "type".
But there have been several instances since the launch of AdSense that many publishers have posted here
