Forum Moderators: martinibuster
Google decides who gets the best position in your AdSense strips by rewarding AdWords advertisers with higher CTR's, with more frequent and better positioned ad displays.
So if an advertiser is good at writing ads that attract clicks, but don't convert well, smart pricing takes it out on the publisher. If your page(s) happens to attract a large amount of these kinds of advertisers, your EPC will nosedive. (I would guess that the Ebay and other similar broad spectrum ads would fall into this profile).
To be fair, AdWords program would need to start taking into account how well an ad converts before promoting it in the ad display queue. This won't happen many advertisers won't use Google's conversion tracking tools, for fear of giving away sensitive information to Google (this has been said on a few of the AdWords Forums).
Just a thought.
However, Google is a little vague on the exact way smart pricing works. But here is something to comfort you. If an ad does badly becuase the ad is badly designed:
1. It probably won't last, because of the minimum CTR rule for Google Search.
2. From what I can see, smart pricing adjustments are relative. All sites are going to be affected by the same poor conversions, so your site is not going to be worse off.
Not discounting the chaos theory either or the fact Google has an interest to maximise revenues.
But if your site (or a section/ category) attract *lots* of people writing good ads won't the following be true:
a) they will continue to bid against each other on price (ok so perhaps a little lower return, but if there are lots then supply and demand has to kick in)
b) Even if (a) is not true, If lots of them write such good ads that attracts many clicks then all displayed Google Adsense Ads will be compelling and therefore more likely clicked on.
Result: higher click thru rate to offset potential lower EPC.
In other words, I'm not too worried. Could affect lower trafficked sites? Possibly. Or those already at optimal clickthrus - Maybe.
Just my $0.02.
I don't mean to be argumentative.
And to proove it, if I'm missing something here like is Google keeping a higher share? then please feel free to rub my nose in it. :)
P.S. I love this site, can't believe I just found it! A real gem. A well moderated, professional forum - I was beginning to think they were extinct.
(I wish I could remember and share where I found it)
As far as I am aware there is no solid evidence that smartpricing is affected by conversion rate. Google have never said that. It is just a theory put forward by some here.
It might be a valid theory, it might not.
All Google have said is that smartpricing works on how valuable they think a page is likely to be to the advertiser. Conversion rate would be one way of judging that, however since it is totally dependent on advertisers using the Google conversion tool it would probably not be very reliable.
The only smartpricing example I've seen from Google talks about the difference between a general page about digital photography and a page reviewing a specific digital camera. My reading of this (possibly wrong) is that smartpricing is based to a large degree on how Google analyses the content of the page; a page they think more likely to appeal to someone in "buy" mode gets rewarded.
Given the mess that is Allegra, Google's current ability to analyse a page is not something in which I have a lot of faith.
All Google have said is that smartpricing works on how valuable they think a page is likely to be to the advertiser. Conversion rate would be one way of judging that, however since it is totally dependent on advertisers using the Google conversion tool it would probably not be very reliable.
It would be a lot more reliable than basing smart pricing on guesswork alone, and Google must have a huge quantity of conversion data to use in determining the likely conversion rates of different types of content (or, conceivably, of different sites).
Of course, Google isn't telling how it determines smart pricing, and that's probably just as well.
As for Allegra, it's had a very positive impact for the keywords that I monitor and for the searches that I've conducted since the update. As the expression goes, one man's meat is another man's poison. :-)
I have emailed G about it and this is what they said.
<snip>
This is a load of crap. I expect some fucuations but not a 90% drop.
Fluctuations means up and down.
As far as I have seen it is just going down down down......
At this rate it is not even worth fooling with.
[edited by: Jenstar at 8:49 pm (utc) on Feb. 18, 2005]
[edit reason] No email quotes as per TOS, thanks! [/edit]
