RSS
Twitter
Oh Look, Yet Another Attempt At Cost Per Action Advertising
from the haven't-we-seen-this-before? dept
After every media mention of "click fraud" on Google or Yahoo, you always hear someone suggest that "cost per action" advertising is the solution. Google and Yahoo, of course, work on a "cost per click" system, meaning the advertiser pays each time someone clicks on the ad. That's where clickfraud comes in -- since each click costs the advertiser money or makes a publishing partner money. Cost per action, on the other hand, gets rid of that problem by saying the advertiser only pays money if some sort of "action" is taken (usually a purchase, or at the very least filling out a form). The thing is, this isn't even remotely new. One of the earliest commerce trends online was the idea of the "affiliate program" which really was nothing more than cost per action advertisements. You put up links (ads) for books on Amazon and you only got paid when someone took an action (bought the book). That's why we tend to be skeptical each time some new company seems to think it's come up with something new with yet another cost per action advertising model. The latest one is getting lots of attention because it's run by a former CEO at Altavista, one of the many search engines Google crushed, only to have its remains later (much later) snapped up by Yahoo. As for the claims that this will take on Google by being "different," remember that Google has been testing its own version of cost per action advertising already anyway. Cost per action advertising certainly makes sense in some cases, but it's a bad deal for any site if the advertisers just use it for brand building. That is, if they don't actually encourage the action, but just get their name across for a future purchase, then the advertiser gets to advertise completely for free. That's great for the advertiser, but raises plenty of questions about how strong the business model is for the site offering such ads.
Reader Comments
(Flattened / Threaded)
A good solution could be a hybrid system. For example, instead of a flat 10 cents per click charge 2 cents per click plus 25 cents per action. Those prices could be way off, there ought to be a pricing somewhere that will appease both advertisers and google. The key is to make the real money off of the cost per action, and keep the cost per click just high enough to prevent abuse by advertisers, and low enough to entice new business and minimize click fraud.
Maybe I should patent the idea?
(reply to this comment) (link to this comment)
Re:
Well you have a year to patent it since you just publicly disclosed it and you can forget about obtaining any international patents you just lost the ability to do that because of this public disclosure. An who cares if it is not obvious that does not seem to matter these days.
But I do agree we tend to think of things as mutually exclusive when combining concepts often leads to a better solution.
(reply to this comment) (link to this comment)
Niche Markets
I used to get $2 bucks a resume on montster.com. Loved cost per action back then.
(reply to this comment) (link to this comment)
look at it this way: Cost poer action amkes sense. Therefore, it is reasonable for a company to tyr to insitute it. And like any company doing anything they have to try to hype it up. It's not really all that silly.
(reply to this comment) (link to this comment)
A proven way to control those who "just (want to) get their name across" is to require a certain amount of revenue production from advertisers.
This has been common in the online adult industry for years.
I'm sure the smart people at Google can figure out variations of this theme that are fair and profitable for both sides.
(reply to this comment) (link to this comment)
But cost per click is so simple, so elegant. You just need a tiny script on your page and a counter. People click and money flows. It's the equivalent of commercial on TV: someone pays to have it aired, I mute the tv and go to pee. In case of cost per click, I click, decide I don't want to buy and off I go for my next adventure.
Also, I'm convinced altavista was dead (or at least moribund) long before google was hip. And I don't think they got crushed, they were fairly deflated.
(reply to this comment) (link to this comment)
advertising models are sub-optimal but stable...
... and for that reason along we should expect them to stay around. The real question is how long will it be until the customers of advertising find a better business model that is technically feasable.
Personally, I think it will be a few years before some next-generation wunderkind stumbles over a better solution that doesn't involve the current web-scraping technologies that for better or for worse are getting swamped by ever growing search results.
(reply to this comment) (link to this comment)
Re: advertising models are sub-optimal but stable.
So, who owns the patent on Cost-per-action advertising business process anyways?
(reply to this comment) (link to this comment)
Add Your Comment