Forum Moderators: martinibuster
As a real time example, Monday is often the best day of the week for my sites but February 28 was the worst Monday income and EPC day in ages. But our traffic was near record levels. Similar situation last Sat too with dismal EPC and revenue.
I have also noticed an all time high record number of PSA's. But had 'good' targeted ads on many sites for a long time but recently the same sites have far more PSA'S than normal.
After a noticable decline from Jan 28 thru about mid-Feb there was a fairly nice improvement for a while (about mid-month) but lately its back off the cliff again!
Anyone else notice this dubious long term (since 2003) and also short term trend (starting about Jan 28 2005), especially past several days? Any ideas why this happens? Has G possibly reduced its payout percentage? Does G perhaps have less advertisors which may be why there are so many PSA ads?
Could this be related to allowing 3 ad units on each page (and more and more of us doing that) vs the past where there was a limit of 1? Has that caused a double whammy of a reduced ad inventory and also caused lower bid prices, resulting in a significant increase in PSA's?
Of course, as a long time (well satisfied and very happy with G) publisher we are extremely worried about this trend. Nothing we do as far as making our websites better, good products/services, content targeting, getting improved traffic and good ad layouts/colors, seems to stop the ongoing decline. Any feedback would be appreciated.
Adwords and adsense are actually not really slick programs at all. They deliver ads that are somewhat relevant (I know that's a debatable point), but as far as how much an advertiser pays (number of clicks) and how much a publisher receives, Google has a secret bunch of Googlers who just toy with the numbers all day and night.
So, if they think what you are posting here is not nice, they may lower your CPM, or start delivering .03˘ ads to your site. On the other hand, if you complain enough - and this happens to me all the time - your numbers will suddenly jump. It never fails... three complaining posts result in an extra couple of .50 or .75 clicks right away.
It's almost as though some guy in California is watching this board, and when he sees me complaining, he figures, "OK, we've caused this guy enough pain for a while, let's click him up."
Since the whole enterprise is cloaked in secrecy, I assert that my theory is as valid as anyone else's and that all this speculation is idle prattling. (yes, I am an idle prattler, especially just before bed).
Party on, Garth...
Whenever my server is down part of the day and I get far less impressions due to that happpening it invariably results in a nice increase in EPC, but WHY, as it would appear to not cause any effect? This anomaly may explain why some members here sometimes say their income is up while others say way down. The ones who say they are doing good I would guess are likely smaller and lower overall traffic publishers.
That has happened several times over past 1 1/2 yrs and just happpened twice last week. Even though server was down for some hrs and total income was down as a result, EPC increased nicely. In the past I have commented it almost seems like G somehow rewards lower traffic sites with better EPC but it declines as traffic goes up over time. Perhaps that is far fetched, maybe not?
Regarding Europe's call to write an ebook. Yes, it would sell but even though my newly discovered amazing secrets are happening even as we speak (and legal to do and permitted by G) everyone would want a refund. Reason is when they ask G about it they will be told absolutely NO, you are positively not allowed to also do what the 'favored' publishers are permitted to do.
Please note I use the word 'favored' for lack of a better descriptive word. I tried hard to use a different word but could not think of one. 'Preferred' could not be used as that would imply only Premium Publishers but these secretive unannounced methods involve publishers of all sizes, small and large.
In the past I have commented it almost seems like G somehow rewards lower traffic sites with better EPC but it declines as traffic goes up over time. Perhaps that is far fetched, maybe not?
Maybe not. Throughout the weekend, I have undertaken a significant redesign of two of my websites. In order to check the results, I had to visit every single page, more than 600 of them, some of them multiple times. My page impressions have jumped 1,000+ per day above normal.
Result? Pathetically low EPCs and CPMS.
Is there a Smart Pricing-like downward adjustment of EPCs and CPMs for inflated impressions?
...Is there a Smart Pricing-like downward adjustment of EPCs and CPMs for inflated impressions?
IMO that is a good possibility and similar things have happened to my network of sites many times involving odd traffic patterns combined with contrary EPC's. It seems the harder I work and more traffic I manage to get and more sites I put online, it results in non-improved income stats (but much better impression numbers).
Not sure if so called 'smart pricing' are the right words. In fact, I am not even sure what smart pricing really is even after reading hundreds of posts about it.
'smart pricing'? . Try 'outsmarting', that's what it should realy be named, this is what it really is, an (almost free) attempt on our hard earned traffic.
Shouldn’t it be the other way around? more traffic more $$?
Not with G$. You see they have their own sophisticated (PHD based) naming convention for all them sucker trap algorithms they keep coming out with.
Rest assured that the Advertiser is still paying the same per click to G$ regardless of all them 'smart pricing' variables, we publishers have to deal with.
Just because you have the traffic doesn't mean your web site will get the best ads!
If your web site has lousy CTR and conversion, Google may pass those primo ads elsewhere where the CTR and conversion are better.
Don't ask me how I know this, living in the Bay Area does present a few sources :)
Ask no questions, I'll tell no lies.
- CTR up by as much as 2% (comparing to Nov, Dec & Jan)
- Impressions steady
- PPC down by as much as 30%-40%
Overall daily earnings continue to move downwards with every passing day. can't say i didn't see it coming though. It was just a matter of time.
Question asked, what about all them new Adwords advertisers signing up with G$ after clicking the "ads by google" link on the ad panels.
Publishers should be compensated for bringing on new advertisers. It would be only fair for G$ to start displaying this link as an aff link and compensate the site owner for new sign ups.
The ppc payout is riddled with speed humps which causes overall earnings to decline, fast. A fair (how naive) company should have compensated for (the almost free now) brand name advertisements they run on millions of sites.
It was only a matter of time for CTR and RIO to be added to the pricing equation. The same works backwards, it is only a matter of time before it starts biting G$ back in the form of less Adwords sign ups, less overall exposure, less overall money moving through the system (as a result of more and more publishers moving to better paying alternatives).
Or so i wish anyway.
< end rant >
Not sure if so called 'smart pricing' are the right words. In fact, I am not even sure what smart pricing really is even after reading hundreds of posts about it.
"Smart pricing" is actually pretty dumb and it's gonna cost Google big-time when Yahoo brings their show to town. If it worked as planned, EPC would fall on lower quality sites and rise on higher quality sites. This doesn't appear to be happening in any way that makes sense, at least on my sites.
The EPC on two of my best sites dropped steadily over time to the point where I decided it would make sense to move to other revenue models for them. After I dropped the Google ads, almost immediately I started getting emails from advertisers wondering why. One of them suggested a deal where they are now "sponsoring" 160 pages of the site for more than AdSense had EVER paid on the entire site.
The other sites where the EPC would have been expected to drop or stay the same are doing better than ever.
"Smart pricing" indeed.
"Smart pricing" is actually pretty dumb and it's gonna cost Google big-time when Yahoo brings their show to town.
Don't know. Something tells me Yahoo ain’t going to be any better. A click from an online page is worth what it’s worth. It is never going to be as good as a click from the SERPs (RIO wise). Thus, ppc caps or price adjustments (RIO wise) will eventually exist on all networks.
The only place to expect some network competition is in the network need for as much brand name exposure as possible (“Ads by XYZ”). The “ads by xyz” generate new sign-ups. So I expect competition to eventually revolve around this issue (in the form of new sign-ups aff compensation model) but I don’t expect to see better ppc payouts.
Hope I am wrong though.
Don't know. Something tells me Yahoo ain’t going to be any better. A click from an online page is worth what it’s worth. It is never going to be as good as a click from the SERPs (RIO wise). Thus, ppc caps or price adjustments (RIO wise) will eventually exist on all networks.
Yahoo doesn't have to be better. They just have to be there. They'll get thousands of publishers making the switch simply because "smart pricing" has driven them there.
"Smart pricing" is worse than useless for its reported purpose because it depends upon statisical analysis on a set of woefully incomplete data.
You're right when you say that a click is worth what it's worth, but it should be the advertisers deciding what clicks are worth, not a broken Google algorithm.
True "smart pricing" can only work when the advertisers are given a lot more control over which sites their ads appear on (or not) and how much they are willing to pay for clicks from individual sites.
You're right when you say that a click is worth what it's worth, but it should be the advertisers deciding what clicks are worth, not a broken Google algorithm.
100% agree. Seem like they just keep trying to refine an already good wine. It will go completely off by the time they are finished tinkering with it.
It should be 100% the advertiser choice of how much he/she is willing to pay/bid per click and not the algo choice (LOL, while the advertiser still pays the same for the click, how clever BTW). After all, who say's all adwords advertisers have the conversion tracker installed on their "thank-you" pages. How can this algo possibly be 100% accurate? based on network history data? I DON'T THINK SO.
As I said earlier they should have called it ‘outsmart pricing’ and not ‘smart pricing’.
You're right when you say that a click is worth what it's worth, but it should be the advertisers deciding what clicks are worth, not a broken Google algorithm.
How would that work? By having advertisers bid separately on every site where their ads might run?
Simply having separate bids for "search" and "content" ads (as some have suggested) wouldn't be an adequate solution, because the variation in conversion rates between content sites could easily be greater than the difference in conversion rates between the search and content networks.
How would that work? By having advertisers bid separately on every site where their ads might run?
Who knows? That's for Google to determine.
What is quite clear is that "smart pricing" doesn't and cannot work in the way that it's intended and it's costing Google some high-quality advertising venues. This will just get worse for them when real competition arrives.
The first rule for getting yourself out of a hole is to stop digging.
Simply having separate bids for "search" and "content" ads (as some have suggested) wouldn't be an adequate solution, because the variation in conversion rates between content sites could easily be greater than the difference in conversion rates between the search and content networks.
This is 100% correct and the exact reason why "smart pricing" cannot work. You can't take conversion data from a handful of sites and extrapolate the likelihood of other similar sites (or pages) having the same conversion rates. The "quality" of sites varies enormously, even within the same niche and with very similar types of content.
A manufacturer of television sets can predict the number of failures per 1000 units of a particular model with a reasonable degree of accuracy without testing each and every television set produced. Why? Because every set is built with the same process using the same parts.
There is simply no statistical way to determine the "quality" of a lead from a click on a website without gathering hard data from that website. It's impossible because there are too many variables.
Should we expect Google (and their algorithms) to be perfect? Of course not. But we should expect them to at least be in the ball park and their implementation of smart pricing isn't.
So if I mention "mousepad" on a page about user interface design, I don't get ads for custom mousepads.
"because the variation in conversion rates between content sites could easily be greater than the difference in conversion rates between the search and content networks."
While this is a good point, I don't see the conversion rate itself being at the true root of the problem.
My viewpoint to the problem is that only the advertiser knows the real monetary value of a conversion.
Smart pricing algorithm cannot possibly take into account the real profit margins of the advertiser, or the real long-term value of a newsletter subscription, for example.
Thus, as only the advertiser knows the real value of an average click from adsense, she will bid only to the maximum value that the click is worth, and not more.
That's why I would like to see ad values based solely on competitive bidding.
Instead of smart pricing, I'd like to see advertisers given more options to choose where their ads appear, and better adsense tools to track their conversions by the referring adsense publishers so that they could direct their resources to the best websites from Return on Investment point-of-view.
G needs to get back to KISS (keep it simple stupid). Too many of these hot shot dot coms went bust because they outsmarted themselves thinking they could game the market with new whiz-bang technology. Almost always a bad move.
Publishers want steady revenues and increased profits for increased work. It seems from all the post here that the harder you work on your site, the less you earn from Google. Pavlov's dogs would not be salivating, nor should we be.
Exactly!
There is simply no statistical way to determine the "quality" of a lead from a click on a website without gathering hard data from that website. It's impossible because there are too many variables.
...?!
Google has a conversion tracker its advertisers can use. If the AdSense on your site converts for them, it records it.
What could be simpler than that?
If your website converts, you get paid for. If your website is NOT converting, SmartPricing pays you less.
If SmartPricing didn't pay you less, you wouldn't see *any* ads on your site because the advertisers would pull their campaigns from the Content network entirely.
This isn't PhD stuff folks. It's really quite simple.
You want to get/keep a high EPC, a high CPM, then build sites that help people make buying decisions. Plain and simple.
Don't build sites about the Russian Revolution and expect it will make you rich! Or add 1,000 pages around the same topic and wonder why things are going down.
Remember what business you're in, with AdSense - advertising. AdSense isn't some 17th Century Patron paying your way to create art.
Google has a conversion tracker its advertisers can use. If the AdSense on your site converts for them, it records it.What could be simpler than that?
True. And that conversion data is very useful for analyzing the value of clicks for that particular site where the data were recorded. It's virtually useless for analyzing the value of a click on the vast majority of other AdSense sites.
You're right, this is advertising, and the advertisers need to be determining the value of advertising on individual sites, not a broken Google algorithm.
As both an advertiser and publisher, I know what works for me much better an any third party could.
True. And that conversion data is very useful for analyzing the value of clicks on that particular site where the data were recorded. It's virtually useless for analyzing the value of a click on the vast majority of other AdSense sites.
When "smart pricing" was introduced last April, Google mentioned that the type of content was a factor, and it used the examples of:
GOOD CONVERSION RATE: clicks from a camera review
POOR CONVERSION RATE: clicks from a page of photo tips
So let's say (just for the sake of discussion) that Google looked for the word "review" or "tips" and adjusted the advertiser's discount accordingly, regardless of whether the page was about cameras, vacuum cleaners, dog shampoo, or mutual funds. It wouldn't need conversion data from sites on all of those topics to make the assumption that a review page was more likely to result in an advertiser-definied "business action" than a tips page was, because its assumption would be based on the type of content, not the specific page.
Similarly, it could make reasonable assumptions about conversion rates for forums, editorial pages, search results, etc. without necessarily knowing the specific conversion rates of every forum, editorial pages, or internal search-results page in the AdSense network. If it did have specific conversion-tracking information for a specific site, so much the better: Such data could be used to nudge the smart-pricing discount up or down from the default setting.
You don't have to accept this if you don't want to; the more important issue is whether Google and advertisers are happy with smart pricing. We can be sure that Google use surveys, focus groups, and other forms of market research to keep tabs on such matters, and it's reasonable to assume that--if there's a mass exodus of advertisers because of smart pricing--Google will take steps to remedy the situation.
Now, some here have suggested that publishers will jump ship to Yahoo if smart pricing isn't replaced by letting advertisers bid on individual sites (or, more appropriately, on individual pages, since conversion rates is likely to vary for different types of content on the same site). This claim is based on at least two questionable assumptions:
1) The assumption that Yahoo won't have something like smart pricing. (It's obviously going to need some kind of bid-adjustment scheme for advertisers, because advertisers won't pay full retail for Yahoo content clicks when Google is offering discounts.)
2) The assumption that Yahoo will pay better if it doesn't have a counterpart to smart pricing. (If Yahoo were simply to use separate bidding for search and content, as some have recommended, the overall result might be less favorable for publishers, or at least for publishers whose pages fare relatively well under smart pricing.)
I don't think it's as simple as you think. For advertisers that use adwords to get leads, what is a conversion? A new client...or a new client that pays off big? I'm sure a few advertisers have figured out that they can potentially lower their adwords bill by not reporting conversions, whatever the he** those are. And many advertisers may simply not want to TAKE THE TIME to report conversions. They have BUSINESSES to run. So, do you really think publishers should have their payouts determined by this kind of system? Do you think Google takes less money based on so-called conversion data? I really doubt it. The only ones who get raked over are the decent publishers while google allows the proliferation of low quality sites to affect earnings for everyone overall. The guy who posted the following quote sums it up pretty neatly.
"Publishers want steady revenues and increased profits for increased work. It seems from all the post here that the harder you work on your site, the less you earn from Google."
True, many publishers are seeing no negative changes and even increases. But those publishers should ask themselves----How much work are they doing to stay in the same friggin' place, or see only minimal increases?
As far I'm concerned, the watchword is "yahoo".
When "smart pricing" was introduced last April, Google mentioned that the type of content was a factor, and it used the examples of:GOOD CONVERSION RATE: clicks from a camera review
POOR CONVERSION RATE: clicks from a page of photo tips
This assumption has several flaws, most important being that it depends entirely on the quality of the review page, which can and does vary by a wide margin from site to site. For example, I have seen pages like this:
Title: Review of Canon ZZZ-21 Digital Camera.
"The ZZZ-21 is a great little digital camera. It takes great pictures and fits good in your hand. Highly recommended."
That's it - that's the entire "review".
Call me crazy, but I would think a high-quality general interest page describing the advantages of digital cameras over film models would convert much better than this kind of "review" page.
It's really quite simple - conversion data from one site (or 100 or 1000) cannot accuratelly be extrapolated to other sites, even of the same type of content.
You don't have to accept this if you don't want to; the more important issue is whether Google and advertisers are happy with smart pricing.
Companies have been known on more than one occasion to be completely happy with policies that in the long run were detrimental to their success. It's Google's ball and they can set the rules any way they want. But if the rules are poorly thought-out and/or implemented they'll be less successful than they would have been otherwise at the end of the day.
The more traffic you get to your pages the less you earn
That hasn't been everyone's experience, and if it were true, there wouldn't be any members of the "UPS Club." Maybe you meant "The more traffic you get to your pages, the lower the percentage that you earn"?
Let's assume (just for the sake of argument) that Google did have a tiered payment schedule that paid X% for the first 1,000 clicks, X-5% for the next thousand, X-10% for the next thousand, or whatever. Would this be inherently dishonest, unfair, or evil? Not really. If Google wanted to encourage AdSense membership by raising the floor (on the theory that the rich are getting rich anyway), then it would have the right to do so, just as an affiliate program that pays a higher percentage to high-revenue publishers has the right to skew its compensation formula in that direction. Higher-traffic publishers might not like it, but smaller publishers would probably be delighted by such a scheme.
Not dishonest. But it wouldn't do a good job of serving publishers who:
1. have high quality content sites that
2. operate in low traffic niches that
3. have a regular contingent of well paying advertisers
Adsense has been wonderful, but it's cookie cutter approach falls short in a lot of way. I think while adsense has provided decent income to webmasters it has also allowed many to realize the near-and/or-future potential of their sites, by seeing WHO the advertisers for their content areas really are.
I think while adsense has provided decent income to webmasters it has also allowed many to realize the near-and/or-future potential of their sites, by seeing WHO the advertisers for their content areas really are.
Maybe, but that doesn't necessarily translate into competition for Google from publishers. For example, I might see a travel agent's ad for Whatsit River Cruises in a Whatsit River Cruises review on my site most of the time, but unless I can deliver a large number of impressions on that specific topic, it probably isn't worth the advertiser's time to buy ads from me--nor is it worth my time to pursue the advertiser.
As for the claim that competition could hurt Google, who's arguing with that? (Although, as I've said before, Google is unlikely to stand by idly while Yahoo raids the chicken coop. I suspect that Google is biding its time while waiting for Yahoo to make the first move.)
You're right. It would depend entirely on the niche and the advertisers, not to mention the quality and breadth of the site. I doubt if many adsense sites could make the transition to cutting out the google middleman. But some probably have that capability. All the more reason to keep writing content and developing a site to further spread its reach and make even clearer to potential future advertisers the worth of the site.
"Google is unlikely to stand by idly while Yahoo raids the chicken coop"
You would definitely think so. The questions are, how many eggs will they get and how will Y define a golden egg versus a bad one?
If she's bidding x dollars for a keyword on a campaign and the campaign makes a loss, there is only one thing needed to be done for the advertiser.
What is it? She lowers her bid prices to make the campaign profitable.
No smart pricing needed.
For "smart pricing" to be "fair pricing",
1) publishers whose pages convert better (higher than average) should be paid MORE than what the advertiser has bid, and
2) publishers whose pages convert poorly (lower than average) should be paid LESS than what the advertiser has bid
and the price for the advertiser should remain EXACTLY what she has bid.
The average conversion rate (the balancing point) for clicks is a matter of statistical analysis.
