homepage Welcome to WebmasterWorld Guest from 54.166.228.100
register, free tools, login, search, pro membership, help, library, announcements, recent posts, open posts,
Pubcon Platinum Sponsor 2014
Home / Forums Index / Google / Google AdSense
Forum Library, Charter, Moderators: incrediBILL & jatar k & martinibuster

Google AdSense Forum

This 169 message thread spans 6 pages: < < 169 ( 1 2 3 [4] 5 6 > >     
Google AdSense Testing Cost Per Action
alika




msg:1354205
 2:23 pm on Jun 19, 2006 (gmt 0)

Anyone got the invite to test the new Adsense feature - Cost per action ads?

 

gregbo




msg:1354295
 8:19 pm on Jun 24, 2006 (gmt 0)

Maybe not as dependent, but about 40% of their earnings are from the content network -- i.e. AdSense publishers -- (the rest is from the search network). This is quite a lot, when you consider that publishers get about 75% of the total revenues. (In total, the content network yields much more money than the search network.)

Revenues are not the entire story. What is the cost to G to support the content network? To analyze click fraud on it; to respond to challenges on charges; to hunt down and eliminate fraudulent publishers; to defend itself against lawsuits? What are the trends going forward? These considerations will have an effect on the decisions G makes, not just revenues.

I'd be surprised if G withdrew its CPA program based on some mass publisher movement to leave AdSense.

PowerUp




msg:1354296
 11:14 pm on Jun 24, 2006 (gmt 0)

I'd be surprised if G withdrew its CPA program based on some mass publisher movement to leave AdSense.

There's no need to withdraw CPA. Just making the CPA costly will do. For example, minimum $100 per action.

moTi




msg:1354297
 1:32 am on Jun 25, 2006 (gmt 0)

I don't think G is as dependent on publisher revenue as you say. They have lots of ways to bring in revenue.

they have one product that brings in the vast majority of revenue: contextual ads.
they have two major sections in which the money is earned: search and content network, about 50/50 split.
claiming that at this stage google is "not so much" dependent on content network is wishful thinking. nonsense in the ears of a shareholder who expects maximum profit.

I said nothing about the absolute success of AdSense; I discussed how CPA is less risky than CPC or CPM.

advertiser perspective. in publisher perspective, this is exactly the other way around. in terms of risk: cpa > cpc > cpm. time to broaden your mind that advertisers are not the only players to continue this interactive game.

The future, of course, will depend on whether CPA becomes sucessful enough that a majority of advertisers switch to it.

nope. the future will depend on how all program participants are able to monetize on the given accounting method.
ppa is by no means the be-all and end-all. in fact, it may well be that network overall earnings will decline with this configuration as risk distribution is worsened for search network as well as content network.

What is the cost to G to support the content network?

you call content network as it is a financial burden for google? come on.. you really think, that managing ppa fraud control is that much easier? remember financial risk is shifted towards google and its affiliates. that will make monetary damages, this time committed namely from the advertiser side, a lot more severe. think about a black sheep, that cheats the system. take 1 action = 500 clicks. cheating one action = cheating 500 clicks. being a publisher, i shudder to think about this scenario, given the fact that ppa ads produce income like a lottery ticket anyhow (say one action per week). not to mention the tracking difficulties in general.
as i said: risk is completely transferred to the other side. now you understand concerns from a publisher perspective? not your problem? it could be your problem eventually.

I'd be surprised if G withdrew its CPA program based on some mass publisher movement to leave AdSense.

i'd not be surprised if google scores an own goal with the purpose to retrain publishers to salesmen.

John Carpenter




msg:1354298
 10:09 am on Jun 25, 2006 (gmt 0)

Revenues are not the entire story. What is the cost to G to support the content network? To analyze click fraud on it; to respond to challenges on charges; to hunt down and eliminate fraudulent publishers; to defend itself against lawsuits?

You think that there is no click fraud on the search network? Of course, there is. Advertisers are fraudulently clicking competitors' ads in order to waste their budget and edge them out of the ad auction.

ann




msg:1354299
 11:26 am on Jun 25, 2006 (gmt 0)

How quickly people forget how Gator, Ezula and the like just blatently stole our prospects away from us.

Google will have a new battle as scum think up newer and better ways to rob the websites (content network).

Ann

europeforvisitors




msg:1354300
 4:40 pm on Jun 25, 2006 (gmt 0)

Affiliate programs work well for my topic, but I've been burned by several dishonest affiliate merchants and would be very reluctant to risk entrusting my revenues to dozens or hundreds of unknown advertisers.

OTOH, PPA is less subject to fraud (depending on the action).

It's less subject to fraud on the publisher side, but it's obviously far more subject to fraud on the advertiser side.

kwasher




msg:1354301
 4:46 pm on Jun 25, 2006 (gmt 0)

Maybe a compromise?

PPCB?

Publisher gets lower pay per click but gets a bonus for each sale.

caran1




msg:1354302
 4:49 pm on Jun 25, 2006 (gmt 0)

I have also lost money with dishonest CPA merchants and feel the minimum CPA should be sufficiently high - US$50 or so. if users access the internet through cybercafes, how will they track cookies?

dakman




msg:1354303
 5:17 pm on Jun 25, 2006 (gmt 0)

Google's conversion tracker is pretty accurate. Top that off with frequent crawling of advertiser/merchant landing pages and checkout pages to make sure the code is there, I think CPA could be tracked effectively (with minimal fraud).

jomaxx




msg:1354304
 5:39 pm on Jun 25, 2006 (gmt 0)

Crawling won't work. I guess in a few extremely simple cases an "action" could simply be defined as having the user click through to a further-information page, but normally you'd have to test the tracking by manually filling out a form or making a test purchase.

europeforvisitors




msg:1354305
 7:20 pm on Jun 25, 2006 (gmt 0)

I guess in a few extremely simple cases an "action" could simply be defined as having the user click through to a further-information page...

Good point. In current Google lingo, a conversion isn't necessarily a transaction (which makes sense if an advertiser is pursuing leads), so maybe an "action" wouldn't have to be a sale, either.

In other words, it's possible that CPA could simply be a more extreme version of smart pricing, with greater potential rewards for publishers who felt confident enough to take on greater risk.

gregbo




msg:1354306
 8:39 pm on Jun 25, 2006 (gmt 0)

advertiser perspective. in publisher perspective, this is exactly the other way around. in terms of risk: cpa > cpc > cpm. time to broaden your mind that advertisers are not the only players to continue this interactive game.

Advertisers provide the money that pays Google, publishers, and (unfortunately) fraudsters. The less advertisers feel secure in their expenditures, the less money they spend. Thus, it makes sense to argue risk from the advertiser's perspective.

ppa is by no means the be-all and end-all. in fact, it may well be that network overall earnings will decline with this configuration as risk distribution is worsened for search network as well as content network.

I think you should go back and read things I have written on the subject of fraud and risk with regards to cost-per-{discrete operation} advertising. By no means do I think CPA is the solution to all problems; it has quite a few problems. It is, however, less risky than CPC or CPM from an economic and technological standpoint.

as i said: risk is completely transferred to the other side. now you understand concerns from a publisher perspective? not your problem? it could be your problem eventually.

Please go back and read what I wrote above regarding the assessment of risk.

gregbo




msg:1354307
 8:42 pm on Jun 25, 2006 (gmt 0)

You think that there is no click fraud on the search network? Of course, there is. Advertisers are fraudulently clicking competitors' ads in order to waste their budget and edge them out of the ad auction.

I think you should go back and read things I have written over the past few months on the subject of click fraud. I am fully aware of its existence on both the search and content networks, and from the standpoint of the publisher, advertiser, and search engine/ad network.

John Carpenter




msg:1354308
 8:56 pm on Jun 25, 2006 (gmt 0)

I think you should go back and read things I have written over the past few months on the subject of click fraud. I am fully aware of its existence

I didn't (nor do I intend to) respond to what you wrote in other threads. I responded to what you wrote in this thread.

gregbo




msg:1354309
 9:40 pm on Jun 25, 2006 (gmt 0)

I didn't (nor do I intend to) respond to what you wrote in other threads. I responded to what you wrote in this thread.

I neither said nor inferred that I thought there was no click fraud on the search network.

Chico_Loco




msg:1354310
 10:53 pm on Jun 25, 2006 (gmt 0)

In my opinion, Google created a great advertising program/model in the beginning, or at least they were the first to successfully capitalize on it. Most of the things that have been introduced since by Google give the impression that the company is not as creative and innovative as I thought they were in the beginning.

It remains to be seen, but I believe that a CPA model has the following fundamental flaws, which most likely cannot be rectified:

1) The tracking cookies will eventually end up in programs that block tracking, just as happened to CJ & others.

2) Unless Google are the shopping cart system, there is no way to verify the volume of sales. Merchants (as a whole) will invariably seek to corrupt the tracking of the system.

3) Customers that order via telephone (such as myself) cannot be tracked, so publishers won't get fair credit for sales.

4) As a publisher, I'm not interested in giving visitors away in the *hopes* that they might buy something elsewhere. If I'm sending visitors away from my site, I want them to be bought from me (PPC), otherwise I'd rather keep them on my site thank you very much. That's free branding, and I won't be letting others' companies get anything free from me.

5) Refunds & fraud will become a major problem. Advertisers will want the ability to invalidate commission payments for orders that are returned or are found to be fraudulent. Once they have that ability they will abuse it by incorrectly invalidating some of their commission payments in order to minimize their expenditure. This would be impossible to track, unless of course Google were doing the shipping on all items, which isn't even remotely likely.

6) Defining a global percentage as a commission rate is going to be very difficult. If an item sells for $10, I want at least 25%, otherwise it's not worth it. But, merchants of items in the $1000+ range aren't too likely to be generous enough to give a 25%+ rate of commission.

Of course... what do I know!?

hunderdown




msg:1354311
 11:17 pm on Jun 25, 2006 (gmt 0)

2) Unless Google are the shopping cart system, there is no way to verify the volume of sales. Merchants (as a whole) will invariably seek to corrupt the tracking of the system.

I may have missed something, because that point has been confidently made by several people, but I just don't think it need be a serious problem.

I don't see why the only pressure on merchants will be to under-report conversions. As I understand AdWords, an advertiser whose ads are not performing as well as the competiion, based on bid x conversions, risks them ranking lower in the "auction" system for placement in ad blocks, or even having a campaign paused for poor performance.

That's a feature of the current system that keeps merchants honest for smart pricing purposes. Why wouldn't that feature apply to CPA advertising too, and thus help to keep merchants honest?

joaquin112




msg:1354312
 11:39 pm on Jun 25, 2006 (gmt 0)

Maybe Gbuy will have something to do with this. Say some sort of payment through Gbuy so everything is tracked 100% of the time?

Chico_Loco




msg:1354313
 11:44 pm on Jun 25, 2006 (gmt 0)

Hunderdown,

The "smart" ranking system they use may give an incentive for merchantd to be honest, but there is one problem. The whole thing is relative. If there were 10 ads available for a certain topic or keyword - those ads with the highest CTR (or conversion rate in the case of CPA) will be shown highest in order of performance. Let's forget about the pricing for the moment, because a higher bid by a lower ranked performer would obviously change the ranking order of the ads shown.

OK, let's look at the CPA model: Let us now suppose that 1 of the 10 merchants has his "game" perfected and that he converts much better than all of the other 9. Obviously, his ad would be displayed at the top (neglecting bid price). Now, what would you been thinking if you were that merchant?

How far ahead of the competitors is that merchant? Would he still be ahead of them in terms of ad ranking if he were to report only 75% of the sales, for example. So, if only 3 out of 4 sales are reported and he still ranks higher in the ad list (because of his superior conversion rate), then his capitalist mind is oblidged to report as few orders as possible in order to stay at the top, but not have to pay those commissions on 25% or whatever it may be.

Of course that merchant would have to keep a close eye on all of the variables to find the correct % to report. In essence though, by reporting only what % you have to in order to beat the "smart" system, you can stay on top of the ad inventory and avoid paying some commissions.

I don't think I explained that quite a clearly as it is in my mind, but there's a way to beat the system, for sure, and it would be dead simple to code.

europeforvisitors




msg:1354314
 12:53 am on Jun 26, 2006 (gmt 0)

I think it's important to remember that:

1) Google experiments with a lot of things, and it's a bit early to get worked up over a limited beta test.

2) Since Google gets a cut of publisher earnings, Google isn't likely to offer CPA pricing without adequate safeguards.

3) If CPA is offered, it's likely to be just one more option, not a replacement for the core AdSense CPC product (which is making a lot of money for Google, publishers, and advertisers).

annej




msg:1354315
 1:17 am on Jun 26, 2006 (gmt 0)

I just hope it is something we can opt in and out of. I'm not sure which I would do. I'd have to try it, but I want to have a choice.

My immediate impression is that I already have affiliate ads on my site and they are the ones that I've selected because they are best matched to my topics and ones I trust. So I'd have to see that AdSense CPA worked as well.

gregbo




msg:1354316
 2:41 am on Jun 26, 2006 (gmt 0)

Maybe Gbuy will have something to do with this. Say some sort of payment through Gbuy so everything is tracked 100% of the time?

Here is a link to a Usenix research paper [usenix.org] that discusses methods for minimizing the advertiser's ability to underreport the number of referrals it gets from publishers. To deploy such a method requires a considerable amount of trust between advertisers and publishers (not to mention the SE/ad network).

hunderdown




msg:1354317
 2:48 am on Jun 26, 2006 (gmt 0)

Chico_loco, you explained that well. I'd actuallly worked out something along those lines as a possibility, but if that's what you're worried about, I'm just not as worried. We'll have to see how it works in practice, but my assumption is that if gaming the system isn't as simple as just under-reporting, then a sizable percentage of people won't try it.

We'll see.

Chico_Loco




msg:1354318
 3:52 am on Jun 26, 2006 (gmt 0)

It's not that it worries me really, because if AdSense was switched to a CPA only model program, then I'd use another service to server ads, such as the YPN or the forthcoming MSN publisher network.

I just think that this (coupled with the other items I mentioned above, some of which are even more worrying) generates enough hypothetical problems such that publishers shouldn't participate, or at least be very reluctant to do so.

Of course, I am wise enough to know that we will have to wait and see what comes of this before completely shooting it down. It may be even more profitable for publishers than the system we use now, but I do doubt that very much.

G'nite!

hyperkik




msg:1354319
 5:03 am on Jun 26, 2006 (gmt 0)

Google's not doing this to lose money. Google is experimenting with the goal of making money. If the test shows that CPA will result in less money for Google (and thus for publishers), Google can be expected to reconsider or retool.

If Google can bring back some of the publishers who left the content network due to poor performance, or don't enter due to concerns of fraud, there will be more advertisers competing for the space. And yes, particularly for small sites, CPA may produce more of a revenue rollercoaster, but on the whole revenues to publishers should go up.

If Google doesn't enter this market (assuming it can be profitably incorporated into AdSense) it gives up a tremendous share of online advertising revenue. If it succeeds, it becomes that much harder for its competitors to displace it as the market leader.

Hobbs




msg:1354320
 7:14 am on Jun 26, 2006 (gmt 0)

Since Google gets a cut of publisher earnings, Google isn't likely to offer CPA pricing without adequate safeguards

EFV,
CPA ways to game the system are not very different than CPC SmartPricing.

We need not look very far to predict how well those safeguards will work, the issue should be addressed and fixed in the current system first before expecting publishers and advertisers to trust how adequate the safeguards will be.

Powdork




msg:1354321
 8:00 am on Jun 26, 2006 (gmt 0)

We keep mentioning Gbuy. Isn't it possible this entire program is simply a means to promote Gbuy.
'Sign up for Gbuy and become eligible to enroll in our CPA program.'

moTi




msg:1354322
 8:03 am on Jun 26, 2006 (gmt 0)

all advertisers will jump on ppa ads if offered and leaving ppc ads. why? it's simple: because a secure income without marketing risk is better.

believe it or not, but introducing ppa area-wide will be the most significant impact since the start of the program. and it will change your job drastically. the thing is, ppc and ppa are 2 completely seperate businesses. one is marketing, other is sales.

well, how would you describe your self-employment? a publisher, a website marketer? with ppa you will be retrained to be part of the advertisers' sales team. you'll become nothing else than an outsourced salesman. concerning adsense, so far your job was to generate qualified ad clicks by attracting traffic with your content. soon your job will be to actively promote all kinds of products to receive sales commissions, thereby sending off most prospects for free.
one of the points that has already been mentioned in brilliant message #106 is:
there is no way to correctly track sales resulting from your efforts. especially non-internet or different-computer orders and time-displaced purchases cannot be attributed to their origin in large part. branding, personal recommendation, product presentation - everything without being paid one penny.

so you will be cheated, in one way or another. are you ready for affiliate business? i for one will rather remain a marketer and be gone if this happens.

janethuggard




msg:1354323
 11:30 am on Jun 26, 2006 (gmt 0)

I would never get involved with Google CPA. I have been in affiliate advertising for many years, and have been affiliates with over 3,000 companys in those many years. Of all of them, there was only ONE that paid out fairly, and on time.

Years ago I was a 100% affiliate publisher, and after being robbed blind, repeatedly, I am now less than 1% affilate. Then, I only place that one ad, on one site, during three months of the years, when it converts well, and we make some decent money from it. I pull it the other nine months. (yeah it is ebay)

IMHO there is NO way to guarantee publshers get a fair deal. Before I would go with CPA, I would discount my in-house to the bone. I would make more that way, with guaranteed revenue, instead of the gamble that the check will be in the mail from commissions by advertisers who have a revolving door in the marketing dept and $6hr employees who have no interest in doing their job well enough to make a long term trading partner.

In dirty tricks, I had an advertiser owe me $60 for a sale. The day of the sale, I checked the commission rate, and it was 5%. A week later the terms had changed, and it was a flat rate commission of $10, a loss of $50 to me. I complained for weeks to both the program and the advertiser, who was a Tier 1 advertiser, and I got nothing. I cancelled the advertiser, and haven't used the program since which dropped hundreds of advertisers from our site, nearly all who were non-performers with high CTR but low conversion (according to the advertisers). A year later my $10 shows up.

I have better things to do with my time than fight for commissions.

Hobbs




msg:1354324
 11:51 am on Jun 26, 2006 (gmt 0)

I could tell my own Commission Junction horror stories too..

But what about advertisers that are big or public companies, I wouldn't expect British Airways for example to resort to fraud.

Chico_Loco




msg:1354325
 12:14 pm on Jun 26, 2006 (gmt 0)

I wouldn't expect British Airways for example to resort to fraud

Such large companies aren't likely to commit fraud, no, but they aren't likely to move to 100% PPA either because those kinds of companies need to brand themselves reliably, which can be done better with PPC.

This 169 message thread spans 6 pages: < < 169 ( 1 2 3 [4] 5 6 > >
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