Forum Moderators: martinibuster

Message Too Old, No Replies

Google moving to PPA?

does it even make sense?

         

loudspeaker

6:34 am on Jul 21, 2006 (gmt 0)

10+ Year Member



I've read a lot of discussions about Google's eventual move to PPA (most recently on this board in the "Q2" thread) and this is what bothers me: BUT IT HAS BEEN TRIED ALREADY! And it wasn't *that* much of a success.

If people really want PPA, they can go to Linkshare, CommissionJunction or any other of dozens of networks that do pricing per lead and per sale. Why wouldn't they do it?

In short, because it doesn't work (for most sites).

Web sites that can actually survive on PPA basis are few and far between. Imagine yourself shopping for a cell phone. You check a few web sites, compare, think it over, talk to your relatives and in the end you might even go and pick it up in from a store in person! Or call by phone! Or go directly to the web site of the provider! Who gets paid? In short, nobody. And in any case, if you've made 4 clicks to the same provider and ended up ordering next day, how will they ever decide WHICH of those clicks actually produced the sale? Endless questions with the same answer - you, the publisher, get NOTHING.

My opinion is that Google should look in the opposite direction - that is, finally recognize that the only realistic and fair model is CPM (impression-based).

First of all, it's fairer to publishers. When I see those huge wide-skyscraper ads, they DO make an impression. Why are webmasters not paid until somebody actually clicks? This may be an ad for something you DON'T need to click on - like a clothing label.. But you'll still remember the name and the logo!

Second of all, it's LESS susceptible to fraud. It's simply harder to fake 1,000 impressions than a few clicks. Or, at the very least, it gives Google's algorithm hundreds more "data points" from which to try to determine if it's all fake.

The only problem I see is that they will be forced to finally have a normal review process because if the site is doing something funny like hiding ads in a place nobody can see them, it won't work. I think it's just Google's aversion to manual reviews that is a huge factor in this.

What do you guys think? Where does the future lie? PPA? CPM? CPC?

Mistra

4:27 pm on Jul 21, 2006 (gmt 0)

10+ Year Member



I earned nothing with Commission Junction after 500,000 page impressions. If Google move to PPA, it means all my investments gone to the drain. I hope this is not true. Very sad.

celgins

4:48 pm on Jul 21, 2006 (gmt 0)

WebmasterWorld Senior Member 10+ Year Member Top Contributors Of The Month



Over the past year, I've earned a mere $10 with CJ and almost $300 with Linkshare.

While I would love more CPM with Adsense, PPA would drop me WAY down.

icedowl

5:22 pm on Jul 21, 2006 (gmt 0)

WebmasterWorld Senior Member 10+ Year Member



I actually received two checks from Linkshare a few years ago. Each was for exactly 5 cents. Woooo! Needless to say, I wasn't impressed.

europeforvisitors

5:26 pm on Jul 21, 2006 (gmt 0)



CPA can be very profitable--even for an editorial site--if you've got the right topic and audience.

It isn't right for every publisher or marketer, though, and that's why--if Google offers CPA--it will almost certainly be offered as an alternative, not as a replacement, for CPC pricing. After all:

- There's a precedent: Site-targeted CPM ads were introduced as an alternative, not as a replacement.

- Most of the nearly $1 billion in AdSense revenues last quarter came from CPC ads, and Google isn't about to turn its back on a winning formula.

loudspeaker

7:10 pm on Jul 21, 2006 (gmt 0)

10+ Year Member



JUST TO CLARIFY:

I didn't mean to start the rumour that Google is already moving everybody to PPA (CPA). IT IS NOT THE CASE. I meant to comment on the various posts (here and on other sites) and numerous articles in the press all lauding the potential move as if it were some sort of new magical solution to all the troubles with click fraud. Sorry if somebody got confused.

P.S. Still interested in hearing your ideas on long-term strategies for both google and publishers.

moTi

4:29 am on Jul 22, 2006 (gmt 0)

WebmasterWorld Senior Member 10+ Year Member



cpc is the origin of googles success story and will likely be the core product for the future - flanked by minor cpm and cpa options. in order to gain a huge advertiser and publisher base for a healthy auction based market, you have to deploy an accounting method that satisfies both parties to a large extent. cpc has turned out to be the only payment form that attracts participants in crowds. other business models like affiliate programs lack of publisher popularity and therefore low ad space inventory (coverage) what keeps the whole thing from starting off - whereas publishers marketing cpm ads on their own (google cpm is a flop?) mostly lack advertiser confidence in the ad space supplier. so we need a balance, a common denominator where both parties can meet.

in terms of marketing risk, cpc is located right between cpm and cpa.
advertisers' darling is cpa, because it assures secure earnings, low risk and minimal marketing effort. in a cpa environment, the publishers and google play the affiliate role. you could also say, publishers give up their job to be retrained as salesmen and are part of an outsourced sales department of the advertiser. do we want that? no, because

a) we like our job as independent website publishers. if we'd like to be salesmen, we'd be salesmen.
b) we deliver valuable targeted traffic to the best of our ability. our work is done at the point a click occurs, actions that happen after the click are not in our sphere of influence.
c) we are not responsible for crappy ad copy, crappy landing pages, crappy payment options, crappy products - things that bury a promising click and put interested visitors to flight.
d) as has been exemplified by the op, a sale/lead/action can be impossible tracked to its origin.
e) cpa is fraud ridden from technical and behavioral standpoints. there is no way to properly affiliate the sales and one cheating advertiser causes huge damage in terms of hundreds of clicks or ten-thousands of impressions.
f) as most of us have experienced, payout is ridiculous and awfully unstable/inconsistent. marketing risk is completely pushed off to the publisher, so we'd rather leave the program.

on the other side, the direction will almost certainly also not go to the opposite. publishers love cpm ads, generating impressions is easier than generating clicks or more than ever leads or sales - advertisers hate cpm, as long as they don't want to do branding. divergence loss is too large, just like in offline media. the point is, online media means undreamt targeting possibilities, so forget classic mass marketing. internet advertisers want secure earnings through direct marketing instead of uncertain return on investment without control.

do you remember where we came from? before adsense, publishers offered cpm banners on their websites to no avail. no advertiser wanted to buy ad space. then came affiliate programs, exploiting us with cpa banners (these idiots want to pay us branding campaigns per lead, how absurd is that?) so there was no way for publishers to make a living off a website and no way for advertisers to gain a decent network coverage for their advertisements.

and then came google, revolutionized the market with pay per click and bound us together to one happy family shoveling billions of dollars :)

loudspeaker

5:50 am on Jul 22, 2006 (gmt 0)

10+ Year Member



Thanks, moTi. An insightful post.

vik_c

6:32 pm on Jul 22, 2006 (gmt 0)

10+ Year Member



Even if Google introduced PPA ads, I don't see why they would want it to replace CPC ads especially when publishers are happy and there's no shortage of advertisers either. Fraud would be an issue to deal with in any model. Google has various methods to combat fake clicks and other fraud. This should be further fine tuned if it needs to be.

gregbo

2:04 am on Jul 23, 2006 (gmt 0)

WebmasterWorld Senior Member 10+ Year Member



It isn't so much whether or not G "wants" CPA to replace CPC. It is a question of which option offers the best ROI for which advertiser. CPA will succeed to the extent that it offers sufficient advertiser ROI (and least revenue loss for G).

farmboy

5:03 pm on Jul 24, 2006 (gmt 0)

WebmasterWorld Senior Member 10+ Year Member Top Contributors Of The Month



A Google CPA program could offer merchants a means of attracting webmasters who will promote the merchant's site/products on a "commission" basis, without the merchant having to set up and maintain an affiliate program.

FarmBoy

europeforvisitors

7:59 pm on Jul 24, 2006 (gmt 0)



Sure, and it might work if:

1) It were a alternative to the regular CPC pricing (as it inevitably would be)

2) Google could ensure that publishers--and Google--were protected against merchant fraud.

stajer

9:28 pm on Jul 24, 2006 (gmt 0)

10+ Year Member



Thanks for the incite, loudspeaker. Personally, I think cpm is gone - a few publishers keep it around, but few have the demo data and focused measured audiences to be able to charge reasonable rates.

What if this post was made in the adword forum instead of adsense? Would advertisers prefer ppa than cpc? I think so.

The real problem with ppa is the low conversion rate of the "average" website. How many of google's advertisers really know their conversion rate? How many try to raise that rate?

Google is trying to improve conversion before they go to ppa. Google checkout allows them to control the shopping experience. Google Analytics allows them to see conversioon and help webmasters improve those rates.

PPA combined with tools to help webmasters improve conversion is a powerful combo, imho.

europeforvisitors

12:55 am on Jul 25, 2006 (gmt 0)



Thanks for the incite, loudspeaker. Personally, I think cpm is gone - a few publishers keep it around, but few have the demo data and focused measured audiences to be able to charge reasonable rates.

I've found that Google's site-targeted CPM ads offered a great way to increase revenue from pages (such as photo-gallery pages) that don't do well with CPC. Because Google serves CPM ads only it thinks they'll perform better than CPC ads, it's a win-win situation for me.

Also, CPM ads are making a strong comeback in some sectors such as travel, where the demand for quality niche sites with reasonable traffic exceeds the supply.

What if this post was made in the adword forum instead of adsense? Would advertisers prefer ppa than cpc? I think so.

Sure, and they'd prefer free ads even more.

The real problem with ppa is the low conversion rate of the "average" website.

That's only one problem. Others include merchant fraud (both real and perceived), the question of what constitutes a sale (cookie duration plays a role here), and how to track CPA when the merchant is looking for leads rather than immediate transactions. (For some reason, there seems to be a widespread misconception that AdWords/AdSense is used only by e-commerce advertisers. That simply isn't true.)

farmboy

1:50 am on Jul 25, 2006 (gmt 0)

WebmasterWorld Senior Member 10+ Year Member Top Contributors Of The Month



The real problem with ppa is the low conversion rate of the "average" website. How many of google's advertisers really know their conversion rate? How many try to raise that rate?

If set up properly, the marketplace should take care of that.

Suppose Company X and Company Y both sell widgets.

Company X has a great sales presentation and converts well. A lot of publishers would want ads for X widgets and that demand should help X keep their advertising costs down and/or allow them to be selective about which sites get to promote X widgets.

Y on the other hand, with a lousy site and few conversions, would either have to improve the site or offer increasing commissions to publishers to get promoted.

FarmBoy

Erku

2:43 am on Jul 25, 2006 (gmt 0)

WebmasterWorld Senior Member 10+ Year Member Top Contributors Of The Month



I think eventually the supply and demand will balance it. This is a very young industry.

Let's say Adsense, YPN and MSN's ADcontent go to CPA Advertisers like it. But publishers choose not to run them. Advertisers are not getting enough or much traffic, so there will be again demand for CPC.

europeforvisitors

3:30 am on Jul 25, 2006 (gmt 0)



Company X has a great sales presentation and converts well. A lot of publishers would want ads for X widgets and that demand should help X keep their advertising costs down and/or allow them to be selective about which sites get to promote X widgets.

That might be fine for Commission Junction, but AdSense is designed to be a scalable, automated system that allows keyword-targeted ads on tens of thousands (maybe hundreds of thousands) of topics.

Let's use my travel site as an example. I might have articles on hundreds of destinations. Now, the owner of widgetville-rentals.com, an apartment-rental service in the capital of Elbonia, wants to run an ad on the AdSense network. Google finds my Widgetville page, pops a widgetville-rentals.com on the page, and that's that. That's far more efficient for both parties than requiring me and/or the advertiser to study conversion and revenue statistics.

CPA might be workable for certain publishers and advertisers in certain situations. But it won't become a replacement for CPC. (If anything becomes a replacement for CPC, it will be CPM, if only because that's a way of buying ads that mainstream advertisers, agencies, and media buyers are familiar with.)

loudspeaker

7:20 am on Jul 25, 2006 (gmt 0)

10+ Year Member



Also, CPM ads are making a strong comeback in some sectors such as travel, where the demand for quality niche sites with reasonable traffic exceeds the supply.

Excellent point, europeforvisitors! Now, that begs the question: WHAT KIND OF TRAFFIC IS "REASONABLE" for a site to consider selling CPM? Let's say in your area (travel), what is reasonable?

Also, do you think that in a way, Google AdSense is the program of choice for the beginners and people who get over a certain threshold will simply go straight for CPM advertisers?

Case in point: I haven't seen any AdSense on Gawker Media blogs (Gawker, Wonkette, Fleshbot, Defamer, Gizmodo, etc). They do get tons of traffic and I suspect they'll simply scoff at what AdSense would offer them in terms of revenues...

europeforvisitors

2:41 pm on Jul 25, 2006 (gmt 0)



Now, that begs the question: WHAT KIND OF TRAFFIC IS "REASONABLE" for a site to consider selling CPM? Let's say in your area (travel), what is reasonable?

To some degree, it depends on the topic and whether you're selling direct or through a rep firm/ad network that can aggregate impressions. I believe The Travel Ad Network, which I use, has a monthly site minimum of 500,000 impressions, but it can spread a buy of, say, a million impressions across multiple Web sites. (It can also offer site-specific ad flights at premium CPMs.)

Also, do you think that in a way, Google AdSense is the program of choice for the beginners and people who get over a certain threshold will simply go straight for CPM advertisers?

No, I don't, because AdSense is used by any number of huge "premium partner" sites such as NYTimes.com. It isn't a replacement for CPM ads (or affiliate sales, for that matter); it's just another revenue source. For example, on my own travel site, AdSense helps to "fill in the gaps" by providing revenue from topics that don't lend themselves to affiliate sales. Fo example, if I write an article on kayak cruises in Elbonia, it's unlikely to stimulate many bookings through my hotel affiliate partner (which may not represent Elbonia in any case), but with AdSense, I can enjoy revenue from clicks on ads for travel agencies that sell Elbonian kayak cruises.

Case in point: I haven't seen any AdSense on Gawker Media blogs (Gawker, Wonkette, Fleshbot, Defamer, Gizmodo, etc). They do get tons of traffic and I suspect they'll simply scoff at what AdSense would offer them in terms of revenues...

Sure, they offer tons of traffic, but it's unlikely that they'd perform well with AdSense. AdSense ads work best on niche sites (or at least niche pages) where users are researching ways to spend their money. (That isn't to say that direct-response ads can't work in any medium; they can, but not when they're tiny text ads with only a dozen or so words of copy.)

gregbo

9:05 pm on Jul 25, 2006 (gmt 0)

WebmasterWorld Senior Member 10+ Year Member



Let's say Adsense, YPN and MSN's ADcontent go to CPA Advertisers like it. But publishers choose not to run them. Advertisers are not getting enough or much traffic, so there will be again demand for CPC.

It depends on how the CPC is priced. Advertisers (except for deep-pocketed ones) won't go back to paying premium CPC prices knowing that no search engine or ad network can make any kind of guarantee that their accounts won't be drained by fraudulent traffic.

BTW, Microsoft is currently researching a new type of business model called pay-per-percentage, where ads are displayed with some probability.

humblebeginnings

9:21 pm on Jul 25, 2006 (gmt 0)

WebmasterWorld Senior Member 10+ Year Member



Sorry for going off topic, but pay-per-percentage is now presented by Microsoft as a way of online advertising that is immune to click fraud. I think pay-per-percentage just means that the damage of click fraud will be split equally among advertisers.

idolw

10:06 pm on Jul 25, 2006 (gmt 0)

WebmasterWorld Senior Member 10+ Year Member



What if this post was made in the adword forum instead of adsense? Would advertisers prefer ppa than cpc? I think so.

depends what the CPA is?

If my revenue from a singe sale is $10 and Google asks me to pay $15 CPA i prefer to play with CPC.

Also, i am wondering how would that be. The keyword [cheap widgets] would cost $100 per action and [very cheap widgets in the darkest part of Elbonia] would cost $5 per action because it is searched more rarely?
If so, as an advertiser, I need to get a few billion venture capital as it would be the only way to make sure I make a lot of money by buying expensive keywords that are searched very often and enjoy top positions.
And what if I create a website that does not convert and agree to pay a lot? The recent AdWords update shows that landing pages that sell like crazy are not liked by Google.