homepage Welcome to WebmasterWorld Guest from
register, free tools, login, search, pro membership, help, library, announcements, recent posts, open posts,
Become a Pro Member

Home / Forums Index / Google / Google AdSense
Forum Library, Charter, Moderators: incrediBILL & jatar k & martinibuster

Google AdSense Forum

This 135 message thread spans 5 pages: < < 135 ( 1 [2] 3 4 5 > >     
Complete rundown of the AdSense changes
For April 25, 2005

 6:01 pm on Apr 25, 2005 (gmt 0)

Site Targeting

Site targeting is a new way for advertisers to select keywords and sites, and then pick and choose where their ads will appear. From a publisher perspective, if you have a high profile site, or even a quality site within a niche, you could find that advertisers are deciding to target your site specifically. However, those with less-than-quality sites could find themselves with a definite disadvantage.

CPM pricing model

There is a lot of buzz [webmasterworld.com] surrounding the way site-targeting is priced.

With site-targeted advertising, advertisers set a maximum CPM bid - that is, the price they are willing to pay for every thousand impressions and pay on a per-impression basis. This means that, unlike pay-per-click ads, you'll earn revenue each time a CPM ad is displayed on your site.

This means publishers will earn on a CPM basis for these ads. But the obvious concern, especially with a $2 USD CPM minimum bid is that publishers will earn much lower than what they would without this. Here is what AdSense says (emphasis mine):

For every eligible impression, both pay-per-impression ads and pay-per-click ads compete in the same auction. Our technology will automatically display the highest performing ads on your pages.

It is worth noting that all publishers are opted into this.

Expanded text ads

AdSense is testing ads that fill up an entire ad unit, likely to be similar to the way public service ads appear. They haven't updated the ad formats page with the new ad style (hint hint :) ) but an example should hopefully appear here soon:

These ads will currently only be used with site-targeted ads, which will have the added effect of publishers being able to spot advertisers who are site-targeting their sites.

Image ads

Google is adding animated image ads, as well as Flash ads [webmasterworld.com]. I wish there was an option to opt-out of Flash ads, while keeping image ads, so hopefully this will be added in the future.

They have also added the wide skyscraper to their ad units that support image ads. That has been added to the ad formats page, and can be seen here: https://www.google.com/adsense/adformats

They have also now added all the image ad styles into their own "category".

That's all folks :)

[edited by: Jenstar at 6:32 pm (utc) on April 25, 2005]



 10:16 pm on Apr 25, 2005 (gmt 0)

"and 1 choice is less then 4 or 5 ads that appear normally in my formats."

And could contribute heavily to "ad fatigue" which could have an adverse effect on CTR.


 10:23 pm on Apr 25, 2005 (gmt 0)

Let me tell you what will be more interesting than seeing Google CPM ads. We all saw CPM ads and an only an addition of "Ads by Google" under them will be no different than image ads we all have seen before.

If they need to be successful, they have to run image ads competing in a, let's say 160x600 skyscraper unit. Just imagine that. (I am have the copyright owner, by the way) :) Looking at an image ad will personally not attract me, but if I see 2 image ads in a skyscraper between the text ads, than I both look at it, and may click on it. :)


 10:25 pm on Apr 25, 2005 (gmt 0)

If we're all automatically enrolled in this CPM deal, the money from that should be IN ADDITION to any clicks we're already getting, so logic (naturally, applying logic to this is probably a waste of time) would dictate that all else remaining the same, publishers SHOULD EARN MORE.

Yes, that should be true IF the advertisers still have to buy the original cost-per-click deal.

Not necessarily. In my experience of traditional affiliate marketing users almost never click on branding ads.

So if the high paying advertisers are site-targetting the three or four big sites in my field then my niche site could be left with no inventory except low value CPM branding ads that no-one clicks on anyway.

I'm sure Google are reasoning that allowing CPM will bring in a whole load of new advertisers who are scared of PPC. So the increased bidding will push total revenues up.

My worry is that Google's total revenue will go up, the revenue of the minority of big sites will go up - and the revenue for us small fish will disappear.


 10:38 pm on Apr 25, 2005 (gmt 0)

hmmmm.... If we are all automatically optted-in too this thing.... ok....

I don't understand this.

Don't advertizers have to select a site before CPM ads would show on that site?


 10:39 pm on Apr 25, 2005 (gmt 0)

correct me I am wrong. The CPM program could have the effect of driving webmasters to create tons and tons of potentially flimsy pages just to boost the impression count right?

And if google presents a list of sites to an advertiser
for the purpose of site targeting, how the heck is the advertiser supposed to know what a good cpm is?

Will google make some of a publisher's stats available to advertisers? I don't think so, considering that the advertiser could, potentially, also be a competitor of the publisher in question.

Makes me wonder if publishers who currently get $40 effective cpm will be in danger of getting 5 bucks cmp.

Of course, that could only happen if adsense allows an advertiser to opt of the CPC model and only choose CPM. But I don't get the impression that will happen from this:

"This model is different from other CPM models, where advertisers must pay a fixed CPM. Keyword-targeted ads will still follow a cost-per-click bidding model, where advertisers set maximum price they're willing to pay for each click their ad receives."

----whoa!. I just read this: "advertisers will be able to choose site-targeted adwords campaigns as an alternative, or in addition, to keyword-targeted campaigns."

So....you could really get slammed if advertisers abandon CPC in favor of CPM and set ridiculously low CPM rates. Thanks google. I feel so much better now.


 10:48 pm on Apr 25, 2005 (gmt 0)

and where the heck does smart-pricing enter into all this? could we end up the victims of low cpm AND get slammed by smart-pricing?


 11:33 pm on Apr 25, 2005 (gmt 0)

Just got an email about the changes.


 11:37 pm on Apr 25, 2005 (gmt 0)

Wonder if this is nothing more than an enticement to keep advertisors (offering them a CPM model that costs them less) because Yahoo is about ready to start it's program.


 11:46 pm on Apr 25, 2005 (gmt 0)

I think that, on the whole, AdSense publishers will benefit from the CPM alternative. Presumably Google will target the ads by page, not by site - and if you have pages which underperform for CPC ads they are more likely to be targeted with CPM ads which, at least, produce some revenue. For those low keyword bids which are displaced by CPM ads, there should be an incentive to up the bid to competitive levels.


 11:56 pm on Apr 25, 2005 (gmt 0)

Is there no way of opting out of the CPM 'option'?


 12:16 am on Apr 26, 2005 (gmt 0)

I read an article on thestreet.com, lemme see here:

[thestreet.com ]

Near the end of the article, the author mentions Yahoo, and Google's entry into graphic ads, calling it a "direct broadside" at Yahoo. This makes more sense than anything else. As usual, the bit players, i.e., small publishers, are fodder for the beast. This is more about Google maintaining profitability and enhancing shareholder value than something good for publishers.

It will just result in more work for us. Corporations suck the life out of entrepreneurship.


 12:24 am on Apr 26, 2005 (gmt 0)

IMO this would appear to potentially be a significant negative for most of us who specialize in well targeted but low traffic sites which rely heavily on good EPC. The changes would seem to perhaps benefit high traffic sites what with more potential revenue based on impressions, not clicks.

If your site is lower traffic and gets say 20/visits/day but offers a high keyword value service, i.e. mortgage loans, you can still make nice money on PPC.

With this new program it will take you a very long time to make any money since you are paid per thousand of impressions and the example traffic site will take almost 2 mos to get 1000 impressions.

If we assume say a 2% CTR this could mean G may be thinking along the lines that 2% CTR equals 20 clicks/month under PPC., i.e. based on .10 cents/click it equals $2 per 1,000, which perhaps not by coincidence is the minimum bid with the new program too.

These are mostly guesses and assumptions and far from proven to be correct. Any opinions on it?


 12:28 am on Apr 26, 2005 (gmt 0)

"For every eligible impression, both pay-per-impression ads and pay-per-click ads compete in the same auction. Our technology will automatically display the highest performing ads on your pages."

How do they make this determination?


 12:42 am on Apr 26, 2005 (gmt 0)

"IMO this would appear to potentially be a significant negative for most of us who specialize in well targeted but low traffic sites which rely heavily on good EPC."

This is exactly my take on it. And if this is how it turns out, it will be really really really awful.

If you tell advertisers that the minimum bid for CPM is 2 dollars, I would guess that most of them will try to get away with 2 dollars---unless google will have some sort of system in place that analyzes the current EFFECTIVE CPM of the site that the advertiser is trying to target and tells the advertiser that their proposed CPM for Site A is too low. If this kind of safeguard is in place, then an advertiser would be prevented from getting onto a targeted site by way-way undercutting the current ECPM of the site. I would hope that google, with their gaggle of Ph.D's, has built such a safeguard into the system.

The problem with the google adsense email we just got is that it doesn't contain sufficient detail. However, it's hard to believe that google would (I'm praying on this one) would institute something so disastrous for publishers RIGHT BEFORE Yahoo springs onto the scene.

Because I am *#$% sure that if these guys torpedo the earnings of small to medium publishers, you will see them running for Yahoo Publisher Network like a tornado bearing down on a mobile home park.


 12:49 am on Apr 26, 2005 (gmt 0)

I thought this was a good thing.

1. If 1 advertiser really wants to be listed on a specific site, they will need to keep adjusting their CPM until they get past the current eCPM value....am I right?

2. If that happens, they now have their site fully in the ad space. Say another advertisers says, "Hey, I wanna be there" and increases their cost per click etc" I see it growing.

3. It will motivate people to generate more traffic to their site ( if it is low...why not get moving and increase traffic to the site.)

4. Scrapper sites will be history, unless someone finds they are converting well from those sites.

5. Do you REALLY think ALL advertisers are going to opt out of PPC? Those that don't care about branding will still use this model, because text get's the clicks.

I would think that advertisers are going to go for sites they want to appear on. Why not get proactive and make a site they want to appear on.


 12:51 am on Apr 26, 2005 (gmt 0)

Well, Google is superb at making money so I doubt that they'd suddenly decide to essentially give away ads at $2 CPM. There has to be more to it than that.


 1:03 am on Apr 26, 2005 (gmt 0)

" If 1 advertiser really wants to be listed on a specific site, they will need to keep adjusting their CPM until they get past the current eCPM value....am I right?"

I would hope so. But I don't see anything in their email that allows one to conclude that.

To me, this seems to be the crux: If the cpm model is supposed to be an addon, that's one thing. But if advertisers can choose ppc OR cpm, this could be the adsense kiss of death for many small sites. Because why would advertisers choose any sites other than the very best performers, which would only amount to a bare handful of sites?

Scrapers could get scraped out of existence which is not a bad thing, but small no-name mom and pop sites could see their revenue practically vanish if they are in a niche that only has 10 good advertisers and those advertisers decide to opt out of ppc, go for CPM, and do not select the mom-pop site as a targeted site.

Am I looking at this wrong?


 1:18 am on Apr 26, 2005 (gmt 0)

ownerrim, i agree with you 100%. Most companies will pick the CPM over CPC because they pay less money and the ad only shows to certain sites they choose.

If this happens, Google stock will fall like a rock because publishers will be leaving pretty soon.

Why did Google change the current system if they are doing great? I dont get it.

I cant wait for yahoo to come out and allow small publishers to use their ads.


 1:54 am on Apr 26, 2005 (gmt 0)

I'm still a little hazy on exactly how site-targeted advertising will work, but at first blush, it sounds like it could be good for niche publishers with a variety of content and subtopics.

On my site, for example, travel-planning pages tend to earn decent to high CPMs, but image galleries earn miserable CPMs because clickthrough rates are so low. (People who look at photos of London or Paris aren't necessarily planning to travel there.) So, if Google could serve up CPM ads at $2 per thousand on my image-gallery pages, I'd be ahead of the game.

I don't see CPM ads being a threat to contextual CPC ads on my travel-planning pages, because there's too much waste circulation in sitewide CPM ads for all but a handful of my niche advertisers. (A travel agency that's selling luxury dirigible cruises is better off having its CPC ads run on pages about dirigible cruising than spending $2 per thousand impressions for run-of-site campaigns, even if the clicks aren't cheap.)

Bottom line: I'm keeping an open mind, and I look forward to seeing how this works (preferably with ads that are relevant to my site's overall topic, meaning no ads for purple pills).


 2:04 am on Apr 26, 2005 (gmt 0)

...3. It will motivate people to generate more traffic to their site ( if it is low...why not get moving and increase traffic to the site}

LOL, you make it sound so simple! Much easier said than done. For starters, a real good and relevant well targeted domain name is needed which also gets some type-in traffic, and most all of those are long gone.

Getting traffic to your site is one of the most difficult things to successfully do. I am not talking about well targeted low traffic (which many of us now get) and benefits from niche market ppc income. That is much easier than getting the kind of high traffic which could better earn money from impressions.


 2:09 am on Apr 26, 2005 (gmt 0)

Why did Google change the current system if they are doing great? I dont get it.

OK, I don't get it either so let's take it from the top and look at their possible folly.

Some ads have low ROI and perform poorly in the CPM environment so CPC was created to give them better ROI for the dollar spent. Likewise, other ads that perform really well suffer under CPC pricing and the advertisers fare better under CPM. I pushed more than a million impressions last month with a 2% CTR (this month has higher CTR) but at this rate it's a HUGE win for CPC advertisers.

Let's compare the math in MINIMUM bids only:

1M impressions @ $2/CPM minimum = $2,000 [much less than I earned]


2% of 1M impressions = 20,000 clicks * $0.05/CPC minimum = $1000

An advertiser buying $2/CPM would fare worse on this site with these stats as CPC is more cost effective.


However, if you have a 10% CTR as some do:

10% of 1M impressions = 100,000 clicks * $0.05/CPC minimum = $5,000

So an advertiser buying $2/CPM would be more cost effective for this web site opposed to CPC.

If you're ONLY BIDDING MINIMUMS, the minimum CPM rate may WIN over the mininum CPC rate depending on your CTR but as I've shown the reverse is possible that advertisers with high CTR could save money switching to CPM.

Finally, any PSA ads that were previously just wasted inventory and filling the PSA's network wide is a win/win for everyone unless you want the ad to be on target then it's a problem.


This is a serious double-edged sword.

I've showed how some advertisers will make out better with CPC and others with CPM so the web sites with the most to risk IMO are the sites with the highest CTR assuming the advertisers figure this out. If Google was smart (love the PHDs) they would just let people bid to fill in the millions (?) of PSA ads and leave good enough alone otherwise.

It will be interesting to see how it all shakes out but my instincts tell me that unless Google has some controls we aren't aware of that savvy advertisers will figure this new economy out (especially since I just explained it to them) and revenues across the board will drop.


 2:21 am on Apr 26, 2005 (gmt 0)

I'm thinking that sites that have big swings in eCPM from page to page (channel to channel) might take a hit if these CPM numbers are based on a site wide basis.

If it's site wide, then it might make sense to spin off high or low paying sections into new sites.

If the CPM ads are priced according to channels or pages somehow, it might not be so bad.


 2:26 am on Apr 26, 2005 (gmt 0)

ken can you elaborate a little on that?



 3:07 am on Apr 26, 2005 (gmt 0)

Doing the math, if my ads went to CPM, at $2CPM, I would make 1/10 of what I am making now. My CRT is 12%.

For my network as a whole, they would have to pay me $20CPM in order to keep my revenue the same. Thinking that is not likely to happen. My eCPM runs $20-30, network. Many of my pages are running $40-60eCPM.

Looking at where my money is, and how the CPM will affect me, it doesn't look good. I don't see advertisers paying me $20-60CPM.

Even after juggling my ads around, some cpc, others cpm, I can only break even, placing the cpc on my lower traffic pages, and the cpm on my higher. Why go to all that trouble, just to break even?

Thinking on this today, I wondered about the timing. The thought was heavy on mind. Something seemed off about it. But what?

And the question was posed before, why mess with the publishers right when the competition is heating up. Is this their answer to anticipated decline in earnings due to the dog days of summer, and a weakening economy? Forget about the publishers, we need a quick boost in profits for the stockholders, we will kiss up to the publishers later?


is this part of one of those oft whispered Google conspiracy theories, where they lead the small publishers to slaughter, right before they dump the small guys from the ppc programs, offer them CPM only, and keep the big boys on the PPC, thereby getting rid of most of the ppc fraud, that so many people are talking about?

In the course of doing that, they leave only the small, questionable sites in the CPM program, making 1/10th what they had before. The big boys get to reap ppc profits, and in the end, the only one to really profit, is Google. While they can policy-out the ppc fraud from publishers, the ppc fraud from competitors will still be there, but at a much more controllable problem, and much smaller.

Of all the G conspiracy theories I have heard, this one actually makes good sense. I know if I were them, that would be my plan. Could be why it came to me, ya think? It is the timing with the PPC class action that leads to me believe that not everything is exactly as it seems. This single extension of services, alone, will crush the class action. They will have in fact, solved nearly all their ppc fraud, and can say they worked aggressively on it.

The only ppc fraud they could not stop, could not prevent at any level, within the ppc system, could not tech out, the one avenue I discovered, that could have been adding up to millions of dollar each year in ppc fraud... this... will solve, 100%. And by George, it is about the only thing that would have worked, and still keep the ppc system alive. Kudos to Google. They did it. Watch and see.

... My grandma, what big teeth you have.


 3:08 am on Apr 26, 2005 (gmt 0)

I'm thinking that sites that have big swings in eCPM from page to page (channel to channel) might take a hit if these CPM numbers are based on a site wide basis
....If the CPM ads are priced according to channels or pages somehow, it might not be so bad.

Google has said that the highest-paying ads will be served. I assume this is being determined by looking at recent data for each page with CPC ads and comparing its eCPM with CPM bids. If the advertiser's CPM bid is $2 and the page has been averaging a eCPM of $1, then CPM ads get served. If the page has been averaging an eCPM of $3, then CPC contextual ads get served. (At least, I think that's how it works. That's certainly how it should work if Google wants to maximize its revenues and the publisher's.)

So, assuming that I understand things correctly, a site with big swings in eCPM from channel to channel might see CPM ads on the low-earning channels and CPC ads on the high-earning channels.

CPM ads sound like a godsend for publishers with blogs, forums, image galleries, and other content that doesn't attract users who are researching ways to spend their money. But even publishers who normally do well with CPC should benefit by having CPM ads served on the occasional page that averages a low eCPM with traditional AdSense CPC ads. (As I mentioned in a previous post, my image galleries have been loss leaders in the past due to low eCPMs, so I wouldn't be surprised if site-targeted ads gave a welcome if modest boost to my income from those pages.)

[edited by: europeforvisitors at 3:10 am (utc) on April 26, 2005]


 3:08 am on Apr 26, 2005 (gmt 0)


It's the 20/80 principle, sort of. Imagine if 20% of your pages provided 80% of your income.

Splitting off the 20% of the pages into a new site will give the new site a much higher eCPM to work from.


 3:16 am on Apr 26, 2005 (gmt 0)

Site targeting is a new way for advertisers to select ... sites, and then ... choose where their ads will appear.

Is this SmartPricing Version 2.0: Multiple advertisers specifically target, and effectively bid,, for ads to run only on Widgets.com? So, on a website basis, Google can track their profitability? Could this result in a bidding war for space on a particularly well converting site? Presumably such bidding for ad space will profit the specific publisher, if there's domain specific bidding competition?

Is this ROI tracking? Advertiser sends AdVersion1.243 to only to Widgets.com and tracks the return? Sends AdVersion1.456 to SonOfWidgets.com?

Is this the beginning of domain bound, domain limited AdSense? AdSense specifically sent only to chosen domains?

[edited by: Webwork at 3:18 am (utc) on April 26, 2005]


 3:17 am on Apr 26, 2005 (gmt 0)

Splitting off the 20% of the pages into a new site will give the new site a much higher eCPM to work from.

Ken, Google has said: "For every eligible impression, both pay-per-impression ads and pay-per-click ads compete in the same auction. Our technology will automatically display the highest performing ads on your pages." The important word in that sentence is "impression," which suggests pretty strongly that CPM and eCPM aren't competing on a sitewide basis.


 3:28 am on Apr 26, 2005 (gmt 0)

EFV; Yeah, I'm just thinking out loud. Just in case....

I still have too many unanswered questions about this new deal.

What are basically image galleries on my site don't do great, and they are the bulk of my pages, I'd welcome a bump up on those.

Other parts of the site do much better, I don't want to sign up to take a hit on those pages.

If this really does work out to be a page by page thing, there's probably no problem.


 4:51 am on Apr 26, 2005 (gmt 0)

Taking a second look at SiteTarget...

...if majority of our revenues start to come from impressions instead of clicks, many of us are going to move the Adsense blocks out of the premium areas of the webpages that get maximum exposure to the visitors (and most click-throughs).

Making publishers think in terms of impressions will make Adsense more like the traditional CPM banner ad networks, which is not a good thing.

If G would have done this right, they would have given advertisers an easy way to block (lots of) individual websites from showing their CPC ads (just like we have the ability to block advertisers), instead of taking this whole new direction with CPM.

I hope those savvy folks at Y! are not going to make their contextual advertising program this complicated, and instead focus on providing clear CPC revenue sharing rules (defined publisher payout percentage (70%-80%) and NO smart pricing), plus keeping the minimum CPC bids at the $0.10 level.


 5:06 am on Apr 26, 2005 (gmt 0)

janethuggard, i agree with you, if google will force the CPM model, i will only make 1/10 of what i am making now...my CPM is $15 per 1,000 impressions because it is rich in contents and it's targetted. But i suspect the advertisers dont want to pay that much. Google is even setting the minimum to $2 CPM as what i've heard.

Even if i have to modify all 5 sites, i would never be making close to what i make now. Google is really crunching our small income.

However, this will fires back to them, because many publishers will start jumping off the boat if this one doesnt work leaving Google losing some revenues.

I hope this is just an option for us to use it or not.

[edited by: asianguy at 5:52 am (utc) on April 26, 2005]

This 135 message thread spans 5 pages: < < 135 ( 1 [2] 3 4 5 > >
Global Options:
 top home search open messages active posts  

Home / Forums Index / Google / Google AdSense
rss feed

All trademarks and copyrights held by respective owners. Member comments are owned by the poster.
Home ¦ Free Tools ¦ Terms of Service ¦ Privacy Policy ¦ Report Problem ¦ About ¦ Library ¦ Newsletter
WebmasterWorld is a Developer Shed Community owned by Jim Boykin.
© Webmaster World 1996-2014 all rights reserved