Google PPA = Google GYMA

[GuyFromChicago]

GYMA = give your money away.

I mentioned the stats in my blog - in short all the tests I've run using Google PPA have been miserable failures so far. There is an unbelieveable amount of fraudulent completed actions. 4 valid leads out of 124 submissions. Ouch.

Is anyone seeing positive results as an advertiser with Google PPA?

[Huligan]

I'm not doing PPA myself, but have talked with others who are. Their results seem to be better than yours. Some of their campaigns do well while others don't. They just concentrate on the good ones just like in PPC.

What are your conversions? Do you have any conversions aligned to online orders?

[GuyFromChicago]

Originally Posted by Huligan

What are your conversions? Do you have any conversions aligned to online orders?

Lead gen/registration right now - product/sales related PPA stuff kicking off next week.

[Huligan]

Please post your results from your product/sales PPA campaigns after they run for a bit. Fraudulent or not, if you can get visitors to enter their credit card info I'm sure you'll take all conversions. I would also like to hear of another PPA experience. I had worried about PPA results for conversions like yours.

[GuyFromChicago]

Originally Posted by Huligan

Please post your results from your product/sales PPA campaigns after they run for a bit. Fraudulent or not, if you can get visitors to enter their credit card info I'm sure you'll take all conversions. I would also like to hear of another PPA experience. I had worried about PPA results for conversions like yours.

Products sales will be a much "safer" venue I'm sure. I'm expecting (hoping) the results there to be on par with other affiliate programs.

[lorien1973]

Originally Posted by GuyFromChicago

Products sales will be a much "safer" venue I'm sure. I'm expecting (hoping) the results there to be on par with other affiliate programs.

The good part of the PPA (sales) will be that the people who lowball prices cannot afford to pay a lot for a sale. So the bottom feeders will be more excluded from the system. The bad part is that the companies who work on extremely thin margins are willing to make a lot less/sale. I'm curious how it'll all work out.

[Huligan]

Originally Posted by GuyFromChicago

I'm expecting (hoping) the results there to be on par with other affiliate programs.

I look forward to reading how it goes for you. Please keep us posted.

[Micromag]

This is very bad.

The problem is that Google is trying to push the affiliate marketing model into his PPC model but is forgetting that there is ton of fraud into this market. (Just see CJ that bans dozen of publishers every day).

The ideal would be that Google has a reversal scheme for advertisers like affiliate marketing has - but don't know if they will add it.

I would do the following:

- Run a Site report and make sure where these leads are from - maybe is just a matter of filtering some publishers.
- Ask Google a refund for these invalid leads (don't know if it will work) remember to provide your report information.
- Be sure that you are not showing the conversion code multiple times for the same visitor (they charge you multiple times for it)
- Assume that you always will have fraud there and decrease your lead value -- you may increase it again as soon you have a better fraud control in a future.
- you may also implement a post processing conversion feature system that will only show the conversion for any visitor from that site as soon you verify that the lead is ok... the system will present the conversion code for the next visitor from that site to ensure to pay commission, but only if you have certified if the previous lead is ok (I know that would be a pain to design a system like that)

[GuyFromChicago]

Originally Posted by lorien1973

The good part of the PPA (sales) will be that the people who lowball prices cannot afford to pay a lot for a sale. So the bottom feeders will be more excluded from the system. The bad part is that the companies who work on extremely thin margins are willing to make a lot less/sale. I'm curious how it'll all work out.

I think that savvy advertisers will look at LTV (if applicable to their business) and use that as the driver for this method of promotion. I know in a lot of campaigns I've worked on losing money on the first sale is the norm - you're paying to aquire a customer and hope to generate a profit on future sales in which the cost of contact is low to non existent.

Originally Posted by Micromag

I would do the following:

- Run a Site report and make sure where these leads are from - maybe is just a matter of filtering some publishers.
- Ask Google a refund for these invalid leads (don't know if it will work) remember to provide your report information.
- Be sure that you are not showing the conversion code multiple times for the same visitor (they charge you multiple times for it)
- Assume that you always will have fraud there and decrease your lead value -- you may increase it again as soon you have a better fraud control in a future.
- you may also implement a post processing conversion feature system that will only show the conversion for any visitor from that site as soon you verify that the lead is ok... the system will present the conversion code for the next visitor from that site to ensure to pay commission, but only if you have certified if the previous lead is ok (I know that would be a pain to design a system like that)

Good suggestions, as always. I was hoping to not have to lock down my form but that will be the next step, in addition to toying with payouts. I know what I can aquire a good lead for using other venues with my existing lead capture process. I'm willing to invest a little more in getting Google PPA to work since I know they'll be a big player (if not the dominant) in the coming years.

[Micromag]

An interesting way to avoid fraudulent leads would be to do not give a clear description regarding what is the lead or where is your conversion code.

For example you could replace your usual lead description from:

You will get paid when a user fills our application form

by something that do not gives a clear description

You will get paid when an important page from our site is visited

That will not affect your conversion rate from legitimate customers but will make more difficulty for fraudulent guys to know where the conversion code is. (a lazy one will not fill the form)

You liked it? Maybe you improve it by providing a false description to you lead like:

You will get paid when a user fills the Contact Us form

But it is not true as your conversion code is installed on another form (the lead form) -- you may got some directly contacts from the bad AdSense guys but they will not get paid.

I don't know if the last one is against TOS - it is not something beauty but I don't think that there is any TOS description against this doing this.