Forum Moderators: martinibuster
What is the real problem?
I did a month long research on the changes at Adsense and what I found, describes the reason for decline in Adsense revenue in some sectors. It was not MFAs, it was not smart pricing, and it was not publisher failure to create a great site.
It was click fraud.
In surfing the web for personal interest, not business, I noticed something a couple months ago. Not having much time then, I back burnered the research, and more recently decided to take a deep look.
What I noticed was this. There were adverstiser I used to see on almost all website across the web, as well in search results, who were now only being seen in Google site search results. Deciding to check it deeper, I took numerous topics and over a course of the month, cleaning cookies every step of the way, and using several computers, I discovered that sure enough, many of those high paying adwords advertisers are no longer having their ads shown anywhere other than in Google search results.
It seems that once Google offered those Adwords advertisers the option of having their ads shown on Google search results only, publishers's site search, or publisher's content pages, many decided it was the best defence against click fraud.
With their ads only showing up on Google search results, that limited click fraud to competitors, and removed the possibility of publisher click fraud.
So, having been armed with this research and knowledge, I took a look at my own sites. Sure enough. Adverstisers who were once abundant on some of my sites, now are only advertising in Google Search, not on my site, nor any other.
Beyond that, some of those advertisers were also advertising heavily at YPN, they are gone from there too. But, at YPN, I have seen some new large advertisers take the place of those who exited. That is a good sign. But, I have only seen that trend at YPN, not Adsense. Long term, there is no saying those new major advertisers will not exit YPN, for the higher safety of the ads on Google site search only, unless YPN offers the same advertiser options.
So, while many large advertisers have exited from publisher's sites, not all have. Still enough have to make a difference in revenue to publishers in those nitches.
This explains why some publishers are viewing others as whiners, because they don't see the problem, yet. It hasn't affected their nitch. But, you can be sure if click fraud continues, it will eventually affect publisher earnings across the board.
I can qualify this by more fact finding. These past three weeks are normally a very active time in nitches that target "back-to-school". It is the time that all advertisers related to that nitch, advertise heavily. Yet, I have tracked quite a few major advertisers who last year were advertising on all publisher sites, and this year only in Google search results shown directly on the Google site.
The "back-to-school" advertisers are many of the same advertisers who advertise heavily in before the end of year holidays. I think this is an indication that end of year publsher revenues are going to be way off in the entire retail and wholesale sectors, compared to past years.
Bottom line is, Google gets 100% of the ad revenue, the advertiser reduces the risk of click fraud, and the publishers have decreased revenue.
When advertisers opt out of ads showing on publisher's sites and publisher site search results, there is almost no way a publisher can combat that. The only thing you can do, that might even remotely help, is to opt in to letting advertisers chose to advertise specifically on your website, after creating and maintaining a good site. It is a shot in the dark at best, and can take a very long time to show any encouraging signs.
There is your smoking gun. Click fraud and Google's response to it, coupled with a settled Adwords Click Fraud lawsuit, brought to court by a Adsense pubisher who was also an Adwords advertiser. I call it the Arkansas Solution, named after the plaintiff's home state.
Did you notice if these who are showing up on the Google serps advertizers were also opting out of the search network as well as content and search for content?
I'm wondering if they were just as put off by the performance/click fraud of the search network.
Changes on ads on your site? Maybe advertisers specifically have blocked your site, and it has nothing to do with pullng out of the content network.
That's just an example.
The truth is there's no data to support your speculations, while there is some hint that the content network is yielding more profit for google than it did before. If there's been such a massive pullout, you wouldn't expect that trend.
But I don't follow certain statements, such as that the problem is "click fraud and Google's response to it". Are you trying to imply that they don't consider it a major problem? If so, I don't agree. If you're going further and suggesting that tolerate click fraud so that advertisers won't use the Content Network and they get to keep all the money, then I'd ask why they bother keeping AdSense going at all.
[edited by: jomaxx at 8:30 pm (utc) on Aug. 30, 2006]
it's a trend that i've been seeing for awhile.
and btw, the settlement is supposed to appealed(?), and something like 500 advertisers have opted out of it entirely, probably waiting until a better click fraud lawsuit comes along.
That's quite possibly true. I've noticed similar for some search phrases.
Point B
Search Advertisers are afraid of Click Fraud
Point A does not offer clues as to WHY publishers left. It only says they left the room. There is nothing there about motivation. You are guessing at the motivation. It is an unsubstantiated claim.
If you want to look at the issue rationally, honestly, and dispassionately, take a look at some of the past posts in the AdWords forum. I did a quick search and found this:
Opt Out or Not to Opt Out from the Content Sites
[webmasterworld.com...]
eWhisper, a mod who works with many clients and is in a better position than most to know what is on advertiser's minds has this to say:
In industries where the payouts are low, content usually converts well and is used effectively.As bid prices go up, quality sites go down, and many advertisers don't want to be associated or fund, scraper sites or 'made for adsense' sites.
They'd rather keep the money that would be spent in this manner (and save hours of time using the excluded sites feature or hunting down ads that may appear somewhere) and do behavioral banner buys or something else worthwhile that's less time consuming and has a similar ROI.
Nothing there about click fraud fears. On the contrary, it's about image. It looks like image even overrides whether or not the scrapers actually convert.
Now read the entire thread. Only one member claims click fraud, and that claim is unsubstantiated. He hadn't checked his logs to verify the claim before posting about it. At the time of posting, he didn't even know where those clicks came from, and whether it was or was not fraud. It was an unsubstantiated claim.
Here is another one:
MFA sites in Google search network
[webmasterworld.com...]
I don't use the content network because the ROI isn't there for my AdWords campaigns (about $80K-$100K per year). How much of this is due to AdSense fraud and how much is due to particulars about my campaigns is hard to tell.
He doesn't know if there is click fraud there or not.
He goes on to say:
When I started with the content network and saw a lot of the crappy sites my ads ran on, I opted out.
Aha... we're back to the way the site looks, i.e. image.
And then he goes on to say he's opting out because:
It's too hard to tell how much profits a lot of these sites either contribute or suck out of my overall income.
He's unsure about his ROI.
1: He objects to the look of content sites
2: He isn't tracking his ROI from the content network
I think those two reasons account for why people back out.
The Trust Factor
And let me suggest a possible third reason: It may be too hard for the average advertiser to trust third party sites- especially after they take a look at the MFAs and Scrapers. It's easy to trust Google Search. It's hard to trust Made-For-AdSense.biz. But that's a gut level response, right? It's not a thoroughly considered decision.
Here is a quote from an advertiser who is sure about his ROI, and is in a better position to comment:
We've had excellent ROI from content though it converts at a much lower rate than search. We just work the bids down until we get a cost-per-acquisition price that we like. But, that's no different than we do with any of our campaigns - all which have somewhat different converion rates.
In other words, he dials down the price of the click until it's at a reasonable level. Smart pricing or smart bidding?
[edited by: martinibuster at 9:21 pm (utc) on Aug. 30, 2006]
Google is probably being a lot more aggressive about reducing fraud than other major media.
On the other hand, there's no question that some larger advertisers are queasy about the, shall we say, look and feel of a lot of the sites where their ads appear. Bigger advertisers are accustomed to being in a "professional" (no offense meant) and, more important, controlled environment. The wild and wooly Web scares them.
The effect of the class action settlement is likely to be about the same as other class actions ... basically nil. It was, after all, a settlement, not a huge damage award based on clear and convincing evidence of wrongdoing.
Personally, I think -- based on no evidence at all -- that Google's smart pricing is effectively feeding more traffic to "better" sites and slowly starving some of the others. If you look at Google's earnings reports, there is no indication whatsoever that advertising is soft, quite the opposite.
Wouldn't surprise me if there are quite a few sites out there experiencing strong AdSense traffic and increased earnings rather than the decline others bemoan.
the content network is full of click-fraud and offers lousy ROI compared to the search network.
This is a problem Google tried to address with SmartPricing, tried to address again by allowing separate Content bids, and yet the problem is only getting worse.
IMO, Google must start letting advertisers select specific websites to place their text ads on. They currently allow site targetting only for image ads, which perform utterly crappy for me, and I suspect many others.
The content network is slowly sinking (that's my guess; only Google can see the real numbers). Google has got to give advertisers the ability to locate and bid on the quality publishers that can supply decent ROI with little risk of click fraud for their text ads.
Why does Google allow site targetting for image ads but not text ads? Is it a technical limitation? Hard to see why text ads would represent any technical challenge over image ads. Clearly, most of the framework is already in place.
The only reason I can think of for this current dichotomy is that Google knows that as soon as advertisers can site target their text ads, most of them will, and Google will have completely insufficient ad inventory for the (tens of? hundreds of?) thousands of publishers in the small end of the Long Tail.
That leads immediately to the corollary thought: Google will continue to disallow site targetting of text ads until the Content Network gets so bad that they have nothing to lose. I think they are a long way from that point right now, but clearly we're in a situation with serious market inefficiencies right now, which means opportunities for someone.
The crucial point here is that Google no longer faces just a technical problem (e.g., make SmartPricing way better so Content gives as good an ROI as Search) here. Increasingly, they are facing a marketing problem of convincing advertisers to even try the Content Network. That problem is only going to get worse unless Google pulls something startling and new out of their bag.
Google Checkout might have given them the crucial data they needed to make SmartPricing an order of magnitude more effective. Unfortunately, it appears they botched the rollout of that product, or so say the people I've talked to who tried it (or at least tried to try it!).
They could start the move to Pay Per Action from Pay Per Click, but that would probably represent as big a loss of Long Tail revenue as allowing site targetting of text ads. I'm guessing that's not on the table just yet.
I am sure Real Smart Folk at Google are working real hard on this problem, since it's a direct threat to their core income, and the clock is ticking away relentlessly. I hope they make some progress Real Soon Now.
(Seriously, it's 2006 and you're just now aware of this?)
And let me suggest a possible third reason: It may be too hard for the average advertiser to trust third party sites- especially after they take a look at the MFAs and Scrapers.
Also, many advertisers probably assume that all of the AdWords/AdSense junk traffic comes from the content network, not realizing that the "search network" consists of more just SERPs:
[webmasterworld.com...]
Why does Google allow site targetting for image ads but not text ads? Is it a technical limitation? Hard to see why text ads would represent any technical challenge over image ads. Clearly, most of the framework is already in place.
I have to differ with this statement. I have had (and seen) site targetted text ads on my own sites. It's not just images.
Site-targeted CPM text ads are distinctive in appearance: Only one is displayed per ad unit, and the headline and body text are larger than in standard multiple-ad units.
What Google doesn't offer (and maybe should offer) is site targeting of standard CPC contextual ads, not just the run-of-site, one-per-ad-unit site-targeted CPM ads that exist now. I imagine that's what Ronburk and many other advertisers would like to see.
In quite a few SEM/SEO-oriented blogs I've read, the common response to click fraud concerns is to opt out of the content network. Now maybe those advertisers aren't necessarily victims of click fraud, but they at least realize they aren't getting as good ROI there as on the search network. Click fraud or not, something is clicking on their sites but not justifying the cost of their ad spend.
So now I'm sticking to Google search for higher bids and have ads on search & content networks for very low bids.
So, Janet is quite right, at least partially.
Whatever it is, Adsense is proving to be a tough nut to crack!
IMO, Google must start letting advertisers select specific websites to place their text ads on. They currently allow site targetting only for image ads, which perform utterly crappy for me, and I suspect many others.
You can target text ads, as others have said.
HOWEVER, the whole process of finding and vetting sites to target is so torturous that it's simply not worth the energy and time (and we're talking hours) to make it all work.
Getting that working (and they've tried to improve) would be a major step in pulling in advertisers for content, AND reward quality adsense sites.
Since I can not name names, I can't offer you an advertiser that I know for a fact is no longer advertising anywhere other than in Google search, on the Google site. If you need verification, I suggest you do the same research, with a wide range of nitches.
This is not the case in all nitches, only in some. As to why some, and not others, I would think it depends on how much the top advertiser pays, and how much click fraud he/she perceives in their industry, which could be verified to some degree by ROI on ads shown on all sites including publisher's sites in the past, versus on Google's site only, once they made the switch. Those who saw no difference, may have opted back in to all sites, save those they blocked.
I did see some 'big boys' ads on those publishers sites, so not all had opted out, just some. But, for those publishers who had those 'some', it could have meant a serious decline in their revenue.
Since those publisher sites have ranked for many months/years, some on the web long before Adsense was created, and later added some Adsense to their sites, and have had top advertisers since at least a year ago, when I last took note, to see no ads for those advertisers on these ranking sites, and yet the ads showed up before and after visiting a publisher site.
I checked many sites to see if any publisher had these advertisers ads on their site, and I couldn't find one. That is not an advertiser blocking sites, it is clicking that box that says show the ads on Google search site only, within their Adwords control panel.
To verify, I repeated this with other machines I own, on different isp, while surfing from different geographical locations, (I travel nearly continual)and the results were the same.
Ken B, yes I saw they opted out of publisher search as well. When it comes to click fraud, publisher search results are really no different than contexual ads on the publisher's site.
I am just saying, while most publshers are screaming MFAs and Smart Pricing as their primary villian, they are missing a key element in publisher revenue decline, in some nitches.
Further, I would like to mention to those publishers that don't know, that Adwords advertisers have the otion of setting a maximum bid on their ads shown on Google search results, and a different one on publisher's 'content' sites.
That has to be hurting many publishers, as advertisers are willing to take less risk on potential publisher click fraud, and have substanially lower bids than in the past, for publisher sites, specifically.
Diversificatin of revenue is now more important than ever. Heads up for Q4 not performing for Adsense publishers as well as 2005 Q4.
I am out of this thread. I can only repeat what I have already said, which is not helpful. I urge those concerned with dropping revenue, to do some reseach to see if ads they used to see on their site, show up on any other publisher site, as they continue to show up on Google's site search results.
That will tell you if you were blocked, or simply are part of the wider problem of Adwords advertisers opting out of using publisher's sites to display their ads.
For all I know, city transit agencies jack up the number of miles (and therefore impressions) they claim their ad-bedecked buses travel each day.
Actually, it would be difficult for city transit agencies to jack up their numbers for a number of reasons. One is that (at least in the US), it is a federal offense to tamper with the odometer of a licensed vehicle. This provides a huge disincentive for them to engage in that type of fraud. Another is that route schedules, timepoints, and maps are publicly available. Expected mileage (subject to small errors) can be effectively computed from this data. Now it is possible for transit agencies to claim "exposures" for buses that are just driven around (e.g. en route between the depots and the actual routes), but too much of that would eat into their bottom line due to the cost of fuel, maintenance, etc. So while there are certainly ways to defraud advertisers on city transit, it's much much more difficult to do that than CPC or CPM fraud.
Google is probably being a lot more aggressive about reducing fraud than other major media.
Well, at least they say they are being more aggressive. Time will tell.
The effect of the class action settlement is likely to be about the same as other class actions ... basically nil. It was, after all, a settlement, not a huge damage award based on clear and convincing evidence of wrongdoing.
Again, time will tell. My opinion: the next wave of lawsuits will feature some expert testimony from people who have quite a bit more knowledge of the Internet's history, architecture, and protocols than Dr. Tuzhilin.
Wouldn't surprise me if there are quite a few sites out there experiencing strong AdSense traffic and increased earnings rather than the decline others bemoan.
I don't think anyone has claimed that no one can make money from AdSense. At the very least, there is a serious image perception with it (for very good reasons, IMO).
But I don't follow certain statements, such as that the problem is "click fraud and Google's response to it". Are you trying to imply that they don't consider it a major problem? If so, I don't agree. If you're going further and suggesting that tolerate click fraud so that advertisers won't use the Content Network and they get to keep all the money, then I'd ask why they bother keeping AdSense going at all.
It wouldn't surprise me if this very debate is happening within G right now.
You can target text ads, as others have said. HOWEVER, the whole process of finding and vetting sites to target is so torturous that it's simply not worth the energy and time (and we're talking hours) to make it all work.
Plus, site targeting may not be appropriate in every instance. Example: You've got a mail-order business that sells clothing, jewelry, and other accessories for chihuahuas. You'd like to buy AdSense ads on the major dog and pet sites but not on MFAs, parked domains, scraper sites, gmail, etc. that you don't want to be associated with or that deliver a poor ROI.
Right now, the only way you can select the sites where you want your ads to appear is to buy site-targeted CPM ads. But those are run-of-service ads, meaning that--if you buy a site-targeted campaign on a major dog or pet site--you'll be paying to run chihuahua-related ads on pages about bearded collies, malamutes, beagles, wire-haired fox terriers, mongrels, and every other conceivable type of dog. That isn't likely to be cost-effective if you sell only chihuahua accessories. What you really need is site-targeted contextual ads that will appear only on pages about chihuahuas.
Will Google offer site-targeted CPC contextual ads? Or even site-targeted contextual CPM ads? Maybe, if research shows that there's a demand for site-targeted contextual advertising.
My CPC was under a penny.
Hahahahahahahahaha!
You just have to use your noggin to exploit these things to your advantage...
But those are run-of-service ads, meaning that--if you buy a site-targeted campaign on a major dog or pet site--you'll be paying to run chihuahua-related ads on pages about bearded collies, malamutes, beagles, wire-haired fox terriers, mongrels, and every other conceivable type of dog.
An advertizer can target right down to a specific page if they want to. Apparently they can target ...
a whole site (example.com/)
a section of the site (example.com/chihuahua)
or a specific page or pages (example.com/chihuahua/feeding.htm)
From the Adwords Help Center [adwords.google.com]
Site targeting places your ads on individual sites in the Google content network. Site sections take that one step further by placing your ads on only one section or even one page of a site.
with a banner that blended
Clever (or evil). Maybe I should take another whack at it. But seeing the image ads start out with 2 orders of magnitude higher CPC is a big demotivator.
An advertizer can target right down to a specific page if they want to.
Yes. But let me know when I can target my CPC text ad. That would make Content attractive enough to bring many advertisers back. That would cause an explosion in AdWords managers who specialize in maximizing your profits by finding you the best sites/pages. That would effectively crush many small publishers, as lots of advertisers switched the default state from "put my ads on any site Google sees fit" to "put my ads nowhere that I haven't personally investigated". That would raise revenues for "good" publishers, as advertisers compete to be on the limited "hot" pages on individual sites. That would cause savvy publishers to publish site stats designed to help advertisers locate the best pages for their ads. That would be a very different world.
It is taking a month for Adwords to approve ads for the content network, ads that take 15 minutes to appear in the search network. Because of this the content network is not likely to see holiday, and "back to school" ads. (Even if you write Adwords; "please approve my ads for the content network", you receive a polite email saying; sorry we're so busy please just wait your turn!).
I don't know how the Adwords API and the content network integrate, but automated posting of ads must surely slow down a manual approval process for the content network. Perhaps they are simply overwhelmed! Oct. 1's API rate changes may reduce the workload.
Some advertisers may not even realize there is this enormous manual approval delay.
These huge delays are one reason you may not see many ads, holiday and seasonal, on the content network. I hope Oct. 1 is the first step toward improvement.
Relevant links:
[webmasterworld.com...]
The following was moved out of Adsense, (remember to put keywords at the top of your posts!). API changes.
[webmasterworld.com...]
