Forum Moderators: martinibuster
I guess most of you would say my account has been smart-priced. Now what would be a good strategy to fight smart-pricing? Should I remove Adsense code from pages with low ctr? What else can I do?
I wish I had said that.
And as Scurramunga has said, I've also tried putting ads back on the low-performing pages, only to take them off yet again. There are pages on my site on which it just doesn't pay to have AdSense ads.
This has been a good discussion. I think the CTR evangelists--I put myself in that group--and the CTR skeptics are not as far apart as they used to be.
I think it's important to remember that "smart pricing" (discounts for advertisers) is only part of the equation. What you earn from a click is determined not only by what the advertiser pays, but also by the percentage that Google keeps as its commission on each click. For all we know, Google might take CTR into account when calculating its commission and payout percentage, on the theory that the ratio of costs to revenue is higher on low-CTR pages than on high-CTR pages. If that were happening, low-CTR pages might not pay as well as high-CTR pages (all other things being equal), but the lower payout wouldn't be the result of smart pricing.
That's a key insight. There's more to Adsense pricing than just Smart Pricing.
I agree with EFV in that "If that were happening, low-CTR pages might not pay as well as high-CTR pages (all other things being equal), but the lower payout wouldn't be the result of smart pricing." I'd even agree that it's possible that it may not be as a result of smart pricing.
On the other hand, as an advertiser, poor performance on a page (loads of views but nobody clicking) strikes me as poor investment. Therefore, to my mind poor ctr is an indication of performance that should be taken into account by one of the algorythms somewhere along the lines. Logically that would be smart pricing. If it's fair to use the performance on one page to penalise many other pages that work well, or indeed other sites that work well on the same account doesn't strike me as reasonable. We don't actually know for sure this happens anyway.
I personally think that there exists a smart rake-off algo. Every time I get a surge of traffic with a surge of clicks, earnings increase, but not by anywhere near the increase in traffic. EG when I was placed at no 1 in the serps for a few weeks. Loads more traffic, loads more clicks with the same ctr. EPC plummeted. It happens time and time again. Sharp spike in traffic = Google grab extra profit. If this is smart pricing or another algo who knows.
To get back to if removing low ctr banners helps the bottom line nor not:- without a shadow of a doubt yes it does. Especially ones that get a very high number of views.
instead many things you try to improve ctr (that is "convincing users to click") have a negative effect on click quality.
the more you blend your ads, the more users click unintentionally. click quality gets worse, ecpm and epc going down.
it's these oppositional trends which imo make it extremely difficult to squeeze the best out of your ad inventory.
i for one haven't had much success in improving "click efficiency" so far. the best i can do is to strenghten ctr by placing the ads above the fold, put them right in the users' focus (=quantity) and limit the number of ad blocks and maybe kick out some bad ads (=quality).
i've tried many variations so far, but beyond that, there was simply no way to substantionally boost earnings per click or per impression. not to speak of the temporally distortions, when changes come to effect earnings-wise etc.
if a visitor clicks out of interest (and that should be the case) this is a preselection. why should one click out of thousand visitors convert worse than one click out of ten?
Depends on the type of interest, and on the nature of the audience for that particular page.
On my own site, for example, I have several destination photo galleries that usually do poorly with AdSense, presumably because users with an interest in Widgetville or Elbonia are as likely to be schoolkids or armchair travelers as people planning vacations. Plus, some of those users will have found their way to the photo gallery through image search, which means they're probably more interested in looking than in booking. So a click on an ad for "Tours to Elbonia" or "Widgetville vacations" may be inspired as much by curiosity or fantasy as by an immediate need for the advertiser's services.
That doesn't mean all photo galleries convert poorly for advertisers. I also have some photo galleries for certain cruise ships (e.g., the S.S. WIDGETONIA, and those do quite well with AdSense. Why? Probably because they attract people who are considering cruises on the WIDGETONIA and want more information. (Not too many sixth-graders are looking at photos of the S.S. WIDGETONIA's cabins or jogging track while preparing school reports.)
In other words, audience is as important as keywords, and type of content is important mainly because it influences the nature of the audience (which in turn influences user motivation and behavior).
Does Google's pricing algorithm consider the "likely behavior" factor when determining smart-pricing discounts? Probably, if we're to believe Google's explanation of smart pricing, which discusses types of content and uses the example of clicks from a product review being more likely to convert better than clicks from a page of photo tips.
instead many things you try to improve ctr (that is "convincing users to click") have a negative effect on click quality.
the more you blend your ads, the more users click unintentionally. click quality gets worse, ecpm and epc going down.
I suspect that's what is happening in many cases. WW members who spend time analyzing EPC, CTR, and eCPM are likely to be optimizing their pages for maximum earnings, too. Someone who clearly knew what she was talking about recently told me that I might be able to increase my clickthrough rate by 400 percent if I blended ads in with my editorial content. Maybe I'm too skeptical, but I find it hard to believe that the number of converting clicks would also increase by 400 percent.
