Forum Moderators: martinibuster
I see they are now ranking PPA ads based on performance, even if it's only stars.
The improvements are nice, but PPA is still not there yet in my opinion. I think Google still has a bit of work to do and the advertisers have a lot of work to do.
FarmBoy
So please feel free to experiment with referral ads on your sites and find out what works for you.
Thanks, will try again this week then.
You just have to find a few ads that click with your audience and it should turn out to be some nice extra cash.
The categories are unpoliced. Expect to find realy crap matching - furniture removal and glucose monitoring under 'Entertainment'.
Yeah, I see ads for shoes under "automotive" .. sigh.
Brake Shoes.
Brake Shoes.
I wish. More like "Find the best selection of [footware brand] footwear at [footware brand] [location]"
:D
Releasing this to all adsense publishers but not all adwords advertisers could yield a similar effect. Release this badboy Google so you/we can have a uber christmas:)
In a 4x4 grid, I placed 336x280 ppa ad across every page of my site in one corner, and 336x280 ppc ads in another corner. The other two corners were not changed. Both ppc/ppa ads looked identically the same (colors, links, even advertiser/content, etc.) and I'm sure the user couldn't tell the difference.
Strangely, the result was a very poor click through rate for the ppa ads and no conversion. I was looking for a good 50-50% CTR% mix. My PPC CTR% was larger by a factor of nearly 5. Looking at the Adsense reports page, I noticed that there were ads showing up that shouldn't have been. I specifically wanted the "show only these ads" and yet I was getting ads in the same general category. Im sure this attributed to the poor CTR. Duplicate ads would appear for that ad style and number of links. Perhaps I would try to test out with another smaller size later on. My hunch is that I would just get a higher CTR%, but conversion won't amount to a hill of beans.
But my concerns aren't the above...
Clicking on the referral links, I noticed the landing pages of the advertisers were unfair - they had placed buttons on their pages to have the customer call them. I can understand why those were there though because these businesses relied not only upon web orders but also from telephone sales which I have seen from their websites in the past. But heck if I am going to give free advertising when I referred the user to them.
Now frankly, I just cannot see how Google could possibly enforce this kind of advertising as the number of advertisers in the program increases. More so if it is in the business model of the advertiser to take on both telephone and web orders. Do I really have faith and trust in the advertiser in reading that cookie and following through and using it upon my visitors referral and subsequent purchase? I don't think so.
The only way I can see this working is if those landing pages are placed on Google servers and go through the Google checkout system. The publisher and advertiser must have trust in a source - Google. For this is what we are using the service for, aren't we? And why not force the transaction through Google Checkout to seal the deal?
To add additional resources to the trustworthiness of the landing pages, have a complaint page available where the Google police can deal with the issue. In addition, when the landing page changes and is uploaded from the advertiser it does not go live immediately. A notification is sent to all active publishers who have a link to the landing page. A voting approval of one or more votes would trigger it going live.
Like those ideas?
A problem I noticed when setting up the PPAs:
I selected the whole computer category and unchecked the box that asks G to optimize the displaying of ads. I assumed that it would rotate through the whole list. when i put the code on the pages I only saw the same ad over and over on all pages.
I went back and checked the optimize check box and replaced the code - now I get about 5 different ads. Apparently I can't have it go throught the 100 or so ads in that whole group.
now I get about 5 different ads. Apparently I can't have it go throught the 100 or so ads in that whole group.
When you choose the 'optimize' option, our system preselects and displays a set of referral ads that should perform best on your pages. Sailorjwd, the referrals team assures me that more than 5 out of 100 ads would have rotated through your referral unit, to make sure we weren't omitting potentially very successful ads, but it's unlikely that all 100 would've been shown.
-ASA
Thats not really cheating ... it's the publishers choice to show or not to show ads ...
CCF
Thats not really cheating ... it's the publishers choice to show or not to show ads .
Why would Google risk their reputation and integrity in administrating an advertising network with that line of thinking? A caveat emptor approach is a terrible way to build trust and confidence in a referral network.
Traffic leaks as usual with advertisers having phone numbers on sites and other cheat methods.
IMO, leaks should be considered from a business perspective.
If my goal is a 2% conversion rate and I'm getting a 2% conversion rate, I don't care if some of the traffic I sent to the advertiser ended up making a phone call instead of a web purchase.
How have I benefited if I get mad at the advertiser, stop showing his ads and end up with 0 instead of 2%?
FarmBoy
The improvements are nice, but PPA is still not there yet in my opinion. I think Google still has a bit of work to do and the advertisers have a lot of work to do.FarmBoy
Incredible. You start off this thread complaining that Google and the advertisers have work to do. And then switch gears and say you are willing to let advertisers put telephone numbers on their pages.
You sir, are the weakest link.
If my goal is a 2% conversion rate and I'm getting a 2% conversion rate, I don't care if some of the traffic I sent to the advertiser ended up making a phone call instead of a web purchase.
You forget the other option of choosing an alternative advertiser that doesn't cheat. Even if your goal is 2%, I'm sure you'll take the 3% from the other guy, if all things are considered equal, because he doesn't plaster his phone number all over the landing page and rob you 1%. It's one thing to show it on the rest of the web site, but advertisers should create landing pages specifically for google adsense publishers, ideally. They already get that traffic and branding for free.
I think this is your normal growing pains of setting up a new program like google referrals, and I'm sure in due time it will equal, if not surpass, Commission Junctions, but for the time being, I'll let the other publishers test it for me and I'll stick with reputable advertisers with a track record of honesty and performance on CJ.
I know advertisers have to deal with some junk traffic from shady publishers on Adsense (MFA sites, for one), but the risk is fully shifted from advertisers to publishers in this referrals program, so I think publishers are allowed to voice their concerns, just as much as Adwords advertisers voices theirs. Here's to hoping bad advertisers get smart priced too ;)
All in all the program has its merrits but there are too many bugs in the system to make it useful just yet. Large corporations are staying away for now too.
Other concerns:
- setting up the ad space is a nightmare with the iframe interface. 30 minutes with no search feature and as of now you can't filter out offers that aren't the size you need.
- The reporting is suspect still, several ads I havent chosen are showing up in the reports.
- The ads dont rotate. 95% impressions to one ad, 5% between the next two and no impressions for 4 ads.
All in all I'm happy to be just test driving this for now, if I relied on it for income there would be a strong case for removing it right now.
I'm still not sold on CPA either, giving away free space on my sites + giving away free traffic for no return 95% of the time is too heavy a price.
Whether you check the 'allow google to pick best ad' or not they just don't seem to rotate.
I ended up putting the codes for six ads in and using ASP to randomly select one.
And these selection of ads is totally useless, as are the categorization of the ads. I think ringtones show up in every category.
ended up putting the codes for six ads in and using ASP to randomly select one.
I did the same with PHP, and it STILL wasn't very random - when I went to look at my stats, one or two ads probably showed 90% of the time.
Incredible. You start off this thread complaining that Google and the advertisers have work to do. And then switch gears and say you are willing to let advertisers put telephone numbers on their pages.You sir, are the weakest link.
I'm not sure if that was meant to be a friendly jab or an attempt at an insult.
Regardless, I didn't say I am "willing to let advertisers put telephone numbers on their pages." I have no control over that.
I stated how I make decisions about advertisers that do place telephone numbers on a page.
FarmBoy
You forget the other option of choosing an alternative advertiser that doesn't cheat. Even if your goal is 2%, I'm sure you'll take the 3% from the other guy, if all things are considered equal, because he doesn't plaster his phone number all over the landing page and rob you 1%.
If such a 3% advertiser exists, I'll certainly go with him regardless of whether he has a telephone number on his landing page.
FarmBoy
ie. can I recommend firefox and post the "get firefox with google toolbar" referral? From what I have read, this is allowed...
