Forum Moderators: martinibuster
For example, a click on an ad for digital cameras on a web page about photography tips may be worth less than a click on the same ad appearing next to a review of digital cameras.
[edited by: markus007 at 8:08 pm (utc) on April 1, 2004]
I've just had the worst day to date (04/04/04). It's a pitiful thing.
:(
I think this change is a little premature.
Content advertising is a whole different ball game and should be treated as such. Trying to force the market to make it work like search engine advertising (Adwords) will fail in the long run.
Content advertising should be seperated first from search engine advertising by allowing seperate biddings for each and let market forces decide. When advertisers start to figure out how to market to content traffic effectively then you will see bids reflect. Only after this and a bunch more testing is when Google should look into poor converting sites/pages.
Advertisers themselves should realize that it isn't necessarily the publisher's fault for poor conversion rate. It is largely due to the same marketing strategy being used for content sites as it is for search engine marketing!
Any publisher who has dealt with affiliate programs probably has seen totally different conversion rates for the same products on two different affiliate sites. It isn't the traffic is bad it is how the affiliate site itself converts the traffic. Each has a different marketing strategy.
Do you suppose Google could be tracking coversions via the toolbar (ex. - see if users visit checkout.php)?
I recall seeing that Google will manage your entire adwords campaign FOR you (if you have lots of traffic and plenty of money). Perhaps THIS, along with their big-partner sites, is where they drew their conversion info from.
Content advertising should be seperated first from search engine advertising by allowing seperate biddings for each and let market forces decide. When advertisers start to figure out how to market to content traffic effectively then you will see bids reflect.
That wouldn't work unless advertisers could bid on specific sites or at least specific types of content (e.g., editorial, forums, gmail, parked domains, and whatever else Google defines as "content"). Why? Because the difference in conversion rates between, say, an article on widgets and a gmail e-mail message about widgets is likely to be even greater than the difference between the difference in conversion rates between a widget ad on a SERP and one in a review.
Advertisers themselves should realize that it isn't necessarily the publisher's fault for poor conversion rate. It is largely due to the same marketing strategy being used for content sites as it is for search engine marketing!
"Largely"? Partly, maybe. But it's unrealistic to pretend that there aren't intrinsic differences in value between leads from different media--and it's also unrealistic to expect advertisers to jump through hoops to increase conversion rates from traffic that doesn't perform well and which they'd just as soon avoid altogether.
Google's automated variable-pricing scheme is obviously intended to compensate for the fact that advertisers have no say in where their "content ads" appear. It's as if Google were saying to advertisers:
"We won't let you pick your advertising venues, so we're going to offer variable discounts based on our testing of different types of 'content' media. And because we're crunching the numbers at our end, you can get more value for your advertising dollars without having to invest time or money on research."
For advertisers, that message is likely to be more appealing than:
"We're going to continue charging the same rate for low-value traffic as for high-value traffic (even as we add more low-value traffic to our 'content' mix), and it's your job to figure out how you can turn non-converting leads into leads that make money."
I suspect that most of the complaints on this board boil down to one thing: Publishers aren't happy with a drop in earnings. I can sympathize with that; my own EPC and revenues have taken a hit that's in the -25% range at the moment. But I'm objective enough to recognize that all traffic doesn't have the same value, and that Google can't expect advertisers to pay top dollar for every click on a "content ad" when Google's quality standards for "content" media are minimal and are dropping even lower with the introduction of contextual ads on gmail.
BTW, I think there may be one silver lining to the recent changes from a publisher's perspective: Variable click pricing (and compensation) will discourage the quick-buck crowd from cranking out "AdSense sites" that have no real content and which invite clicks because readers can't find any information on those sites' pages. This should help to prevent dilution of AdSense click inventory for competitive keywords, thereby making it easier for legitimate publishers to earn revenues from AdSense.
BTW, I think there may be one silver lining to the recent changes from a publisher's perspective: Variable click pricing (and compensation) will discourage the quick-buck crowd from cranking out "AdSense sites" that have no real content and which invite clicks because readers can't find any information on those sites' pages. This should help to prevent dilution of AdSense click inventory for competitive keywords, thereby making it easier for legitimate publishers to earn revenues from AdSense.
Well maybe if Google was really serious about making sure that its advertisers were getting people genuinely interested in their ads, they wouldn't allow such sites to run their code in the first place.
This is not just discouraging "adsense" sites. This is also discouraging legitimate content sites which brought Google many thousands of dollars in legitimate revenue.
I would urge publishers to look at this from the ad buyers' perspective. G had to do something to increase the ROI to the next step.
What folks in the industry have been saying is that they are trying to move contextual ads to the next level, closer to the cost of yellow pages where the average lead is worth a dollar. Right now, it's closer to 35 to 40 cents. And, it was going to stay there unless G made some kind of move.
To some degree, Overture/Yahoo is where G is heading. Overture has been more selective in their publishers and, in turn, is thought to be getting a better rate of return for their ad buyers. Who knows for real? Well, this move by G indicates there is something to Over's claims. And, it's just common sense, of course.
I've been working with small and medium newspaper web sites. They works soooooo hard on their sites, but they can't get the ads to cover the investment they've put into it. They're annoyed that these guys working out of a spare bedroom are getting the same PPC as they are.
In the past I've said it's all about traffic. Now, however, it's getting to be more on the quality of traffic, too. (Alas, is news high quality e-commerce traffic? My data says it's not.)
Which means that you've got to get into the business that provides high quality e-commerce traffic.
A great example is the new theatre section of the New York Times. That's high quality, e-commerce traffic. They deserve a higher PPC than, say, a chat room on theatre and entertainment.
Aren't they [Ad-Sonar] still in beta?
Yes, but everything works perfectly as far as I can see. I've even received followup phonecalls from one of their partner managers who is trying to help me finetune my ads as much as possible. That's unheard of with Google.
Yes, quigo's ad_sonar is now working better than google does.
There's an idea...
works soooooo hard on their sites, but they can't get the ads to cover the investment they've put into it...New York Times. That's high quality, e-commerce traffic. They deserve a higher PPC
With respect, how hard they work on their sites is not really a material fact and it's not logical for their expectations of revenue to be based on the effort that went into individual articles. Revenue should be based on what the advertiser is achieving from the traffic the publisher sends. News sites are limited in what they can do and often have very, very low quality traffic (from the point of view of e-commerce). People click a headline and read an article because the subject interests them and not because they are in a buying mood. A lot of (non-news) publisher sites can be geared towards - and the content targeted towards - attracting people in the buying mood doing pre-purchase research. News sites can't do that and in relation to those publisher sites they should earn less.
EFV, you are spot on as usual in most respects. Like you I've taken a hit on earnings but there is no other ad network I can make even the current level of earnings on, so I'll be staying with Adsense. I do believe that your prediction of variable pricing having the effect of diminishing the quick-buck sites is wrong. Some of the quick buck sites are big players and Google earns revenue from them and they still have their poor quality automatically generated pages reaching the tops of SERPS. The variable pricing is still variable at Google's pleasure so there may be some poor quality auto generated pages paying pretty high payouts if Google's algo says they should be getting a bigger chunk. (The reliability of doing this without access to ROI does leave me feeling it's pretty much luck. How can an automated system decide which sites should get more money for clicks? Especially when it doesn't have ROI information? Perhaps that's why content sites like yours and mine are showing a drop ;))
Has this reduction in CPC, as seen by publishers, been passed on to the advertisers?
Can anybody verify that advertisers are paying less for thier ads displayed on content website publishers?
Has this reduction in CPC, as seen by publishers, been passed on to the advertisers?
I've seen a few posts by advertisers who have reported seeing discounts, and I've also seen posts by advertisers who have complained that cheap nickel ads are getting harder to come by. So maybe click prices are edging toward a middle ground: fewer bargains at the low end, but cheaper prices in the middle and top ranges. That's just idle speculation, though, and I won't be surprised if somebody here provides an example that shoots the idea down. :-)
2 sailing sites. All information based content
I have seen the same old ads appear but now at much reduced cpc. My ads are spot on theme wise and always have been. The subject might be "learning to sail" and I will see sailing school ads.
I have seen an increase in ads for sailing related clothing and gifts, which whilst still thematically accurate do seem a little a bit out of place.
(Oh and I've just signed up for ad_sonar)
Judging the quality of the traffic you're getting is key to successfully marketing your product, but I found it incredibly difficult to implement. I was lucky to be working in a software company, so I could steal some programmer time to help me build what I needed.
If Google is creating a system which builds in the value of the traffic, that's an enormous benefit to advertisers. Remember... the clients? A fraction of them have the tools to accurately track the return on investment for the traffic Adsense is sending them.
My EPC is down about 25%, but if that's what needs to happen to keep the whole system working, I'm glad to take the revenue hit. I think we're still a few years away from this maturing, but when it does, I think we'll all be able to bank on how stable this source of revenue has become.
Anyway, raising my volume of traffic is easy. Selling advertising is hard.
From my point of view, CTR has dropped slightly, but total earnings for each day has dropped by about 60%!
a) The savings are passed on to the advertiser.
b) The publishers are informed about how and why their revenue is decreasing, so that they can take steps to increase it. Right now, none of us know anything about improving the "quality" of our click throughs.
Basically what Google has done is changed their CPC program into a CPS (cost per sale) program, but measures it in click throughs *when one has absolutely nothing to do with the other!*
I signed up to be a part of a CPC program. Nothing more, nothing less.
My two cents: Google is doing this as a way to quiet advertisers who don't want their ads appearing in AdSense.
My two cents: Google is doing this as a way to quiet advertisers who don't want their ads appearing in AdSense.
But advertisers can already opt out of AdSense, simply by opting out of content ads. And let's face it: In the case of niche topics, at least, AdSense probably offers better targeting and higher-quality leads than many general-interest "premium partners" do (never mind DomainPark or gmail, which are the biggest reasons for not using content ads).
<added>
Yes, but everything works perfectly as far as I can see
Except their release date - which was first scheduled for last year, then had several delays and is still in what.... beta?!
[edited by: Macro at 4:37 pm (utc) on April 5, 2004]
Posted by Adsense Advisor:
So, you may find that your revenue goes up, goes down, or stays the same. It may take some time to see a pattern emerging.
it appears every single person on this forum has taken a big pay cut.
I've not - reasonable pay increase.
But then its an odd niche I play in, and the improved targeting has probably massively outweighed any cost reduction. CTR is at least 50% higher.
Overall, it seems inevitable that it should go this way.
I wonder to what extent Google realise that these policy changes shape the type and quantity of sites gaming their algo in the future.
Making a rod for their own back or steering us into a corral?
A bit OT but jaxomlotus didn't you post something many moons ago saying that A*sonar would be fully operational by mid Jan? Their plans keep getting put back so much I could be forgiven for thinking they were being run by M*crosoft/MSN ;)
<added>>>Yes, but everything works perfectly as far as I can see
Except their release date - which was first scheduled for last year, then had several delays and is still in what.... beta?!
:shrug: it doesn't matter much to me if it's got a beta tag on it or not. It's still displaying ads and is earning more revenue than Google currently does. Don't get me wrong, I used adsense up until yesterday - adsense paid far more than ad_sonar did until that point. But now it's so lopsided as to be ridiculous. I can't help but wonder what made Google decide tampering with the formula was a good thing. Were advertisers leaving them in droves? If so, how does this help? As an advertiser I don't want to pay less for a "low-quality" click - I don't want to waste any money at all. I'd rather pay for one high click than a dozen low ones. If a site is turning out bad clicks, why not just exclude them from the network? Why waste the time and money of everyone involved? As far as I can see, that just can't be the reason.
So what was the problem with the old scheme? If it ain't broke, why fix it?
[edited by: jaxomlotus at 5:20 pm (utc) on April 5, 2004]
I don't think there's enough data to make any firm conclusions. Thurs-Sun are my lowest earning days of the week in any case.
Compared to my all-time averages, this is how the last four days work out, where x is the average.
Thurs. April 1
EPC 0.68x
CTR 1.00x
EPM 0.67x
Fri. April 2
EPC 1.11x
CTR 2.5x
EPM 2.71x
Sat. April 3
EPC 1.32x
CTR 1.17x
EPM 1.54x
Sun. April 4
EPC 0.54x
CTR 0.42x
EPM 0.21x
I had two days up and two days down.
Now I'll compare those same four days to my March averages.
Apr. 1
EPC 0.96x
CTR 0.85x
EPM 0.80x
Apr. 2
EPC 1.58x
CTR 2.13x
EPM 3.25x
Apr. 3
EPC 1.88x
CTR 0.99x
EPM 1.85x
Apr. 4
EPC 0.77x
CTR 0.35x
EPM 0.25x
Two up, two down. My average for the four days is up compared to March.
EPC 1.42x
CTR 1.01x
EPM 1.41x
They are also up compared to my all-time average.
That said, four days worth of data isn't statistically significant, so I don't feel comfortable drawing any conclusions yet.
