Google Pay-Per-Action - - Bad news for publishers

[matz]

Pay-Per-Action (beta): http://services.google.com/payperaction/I think it's potentially bad news for publishers. 1 - The conversions on PPA will initially be much higher than PPC (First-tier affiliate programs, like AdSense, claim an average conversion of anywhere from 8-10%.)2 - Therefore, the bids on PPA advertising models will have to be much higher than PPC, and while this lead to higher conversions. 3 - With much higher bids than PPC, the conversions for PPA will fall. Why?Say advertising in Google gets 1 sale out of 100 clicks - in other words, advertisers get 10% conversion with Google. With a CPA model, publishers won't be paid for the 100 clicks, but only for the 10. Therefore, for publishers, the bids should be 10 x as high as they are now.4 - Will they be? Will advertisers paying 15$ for a click now with PPC simply switch it up and pay 150$ in PPA? Probably not, despite the initial conversion data. So most publishers won't go for this model (if Google gives them a choice, which they surely will, or lets them have both PPC and PPA, which they surely won't) and the ones that do will probably try and cheat it, leading to lower conversions and tons of clickfraud.5 - Clickfraud is estimated anywhere from 15%-40% (my own personal estimates are quite higher). Imagine the clickfraud on an ad valued at $150! Conversion rate and bid fluctuations over any advertising model (PPA or PPC) leaves the door is wide open for fraud.6 - In PPC, a click is a click (unless it's invalid, and trust me, its the advertisers who get the last word on whats invalid, not Google). In PPA, who gets the last word on what an "action" is? 7 - I believe, though I'm not sure, that users can define what an "action" is - who's to say they won't define a click as an action, thus making all the PR Google is telling their advertisers about the benefits of PPA fluff?8 - And what if the advertisers get to decide what an "action" is, thus deciding which clicks they'll pay for? Alarm bells should be ringing for publishers everywhere.9 - Higher bids do NOT mean less clickfraud. In fact, I see more opportunities for fraud (both by publishers and advertisers) with PPA. I think the PPC model works best - it's transparent, earns solid results and revenues, and is in fact easier to monitor against fraud. It also compliements what advertising is supposed to be - a message from the firm to the consumer.Advertising is not just about earning a sale for everytime someone sees your ad. Advertising creates brand recognition, exposure, and more often than not lead to indirect sales. Paying for every click to their website costs advertisers more, but its part of the WHOLE buying process. PPA is like saying to the TV company "ok we're not going to pay you for airing our commericial, we'll just pay you everytime somebody buys our product and mentions your TV station".I started a thread about this in "Advertising, PPC, Affiliation", but from an advertisers perspective.I was approved early for a few of my clients, and have gotten a few sales in the past week, not much though, as it's still in beta and to new.I guess this is not good for a lot of publishers out there, but good for a lot of them. I not only work on this for advertisers, but I am a publisher too, who brings in a few grand per month. But what I see is that my affiliate marketing campaigns (pay per sale), bring in much more money than my pay per click campaigns do (from a publisher and advertiser perspective).Overall I think this is just a better system for advertisers and publishers. It will push publishers to put more unique content and good sites. We should see less of these crappy unrelated Adsense sites. It's also a lot harder to fraud a pay per action (well sale based action) than it is a pay per click system, so I would expect to see less fraud for advertisers. Also, the cost that advertisers pay for a sale will be much more effective for them, because they are setting a cost per sale/lead, and only paying for the sale. Google Adsense/Content Match just sucks for most advertisers, it just doesn't convert...more money for publishers, less for advertisers.Probably much more I could say...anyone else want to chime in. I have setup this program for 3 clients now, so if you have any questions, please ask.The main difference CPA has over PPC is the fact advertisers are more in control and would be able to set actions (not sure what conditions etc will be necessary - but they can gain more control over their marketing spend).Email sign-ups, clicks(visits) and purchases would all be actions - with the purchase action accounting for the majority of any revenue that a publisher would receive.It is bad news for a lot of publishers - those who are not focussed on any commercially appealing content sites that earn a trickle of adsense income over time. However, it will (or should), significantly increase the revenue for the more commercially focussed publishers.Overall, it has advantages and disadvantages with clickfraud playing a less significant role in CPA as the majority of the revenue from CPA is focussed on the actual sale rather than the visit... so the percentage of clickfraud may still remain the same, but the revenue lost due to those activities would be significantly lower than on a CPC model.Ganceann wrote: