Low convertions == Low earnings per click?

[Join us @ facebook]

I've just found this very interesting post regarding the low quallity advertisers for the adwords program.

As Daniel sais,

What smart pricing does is take into account how well clicks are converting, in other words how well clicks result in ?customers? for the advertiser. If clicks from your website convert well you get paid more and if they convert bad you get paid less. The algorithm Google is using is much more complex than this but the basic function remains the same. Smart pricing is said to be determined on a weekly basis so it can cause great variance in your earnings from one week to the next.

Smart pricing is a way for Google to determine the quality of your website and your traffic from the advertisers point of view, but unfortunately the effect of smart pricing affects your entire AdSense account. If one of your websites have a low click to conversion ratio it will cause all of your websites to be smart priced, and there is no quick way for you to know which website is at fault.

It seems that adsense algorithms in addition to a web site that is not converting well, can really slow down your earnings.

Very interesting thoughts, but actually I am not sure how true this is. I guess an expirement could help, but obviously it would be very difficult to do so.

One of my observations is that when I added a proxy website to my network, the clicks raised by 40-50% while the earnings are almost the same. Maybe this has to do with the things that Daniel wrote.

What do you think?

I don't think that this is true, I have really different CTR's with all of my websites and the earnings for each site are very different.
I think it's more about the number of page views and the quality of ads on you site. If you site is showing mostly expensive ads you earn a lot and otherwise not.

True, but the quality of ads shown on your page probably depends on the conversion ratio from your site. So higher conversion --> better ads --> more revenue.

True, but the quality of ads shown on your page probably depends on the conversion ratio from your site. So higher conversion --> better ads --> more revenue.

thats possible, but strange enough I have days that I earn 50 cent a click on a page and sometimes only 3-5 cents (there is not really a structure).

But I think it's possible if you have more then 100.000 views a day that you get (more better ads)

So, how exactly does G count what converts.

Do the advertisers tell G, hey I got a hit and I sold $100 of stuff? Or is G counting a click on my site, takes the visitor to another site where they click another ad? If I click on an ad and visit 5 pages of a site? 10? If I click on an ad and walk away from my machine and 'stay' at the site for 10 minutes? What if I call the company to place my order?

I admit there's a lot of web technology I don't understand, but not sure how this can be effectively measured.

I've seen this topic before and while everyone wondered if they were penalized, no one came up with a way for G to count/define what converts. I keep wondering if this is just more urban mythology.

I think that they can measure a piece of convertions using the google analytics program.

[Join us @ facebook]

I think that they can measure a piece of convertions using the google analytics program.

That's corect. Of course it is really difficult to automatically find out the convertions that a site produces, except for the advertising clicks (I mean the clicks to the ads inside of the advertiser's site)

If you are an advertiser, do you have to use analytics? Isn't that by invitation only?

[Join us @ facebook]

No it is not by invitation, but you are right.

As analytics is not something that you are forced to use as an advertiser, obviously those convertion data should be very innacurate.

Google probably tracks conversion only for advertisers who are willing to participate..

Also conversion may not necessarily mean a sale. It could be a request for more information, reading an article etc. So this also brings in more inaccuracy to the data.

Again this is probably the only way to track conversion rather than not tracking it at all.

[Join us @ facebook]

Well it seems that we were wrong...

Google is launching an optimizer, which will track convertions for adwords advertisers.

That means that this theory will be able to work soon....

Another thought about this last night.

I get a targeted visitor via G, due to my content G serves an appropriate ad, my visitor clicks on it and goes to the most crappy site ever and quickly leaves.

Under what circumstances would it be fair if I am held accountable for the owner of that crappy site not seeing conversions?

It seems under this concept the burden of making the sale is moved to the advertiser rather than the seller. If that's the case, I think advertisors should get more than 200 competitor bans.

Another thought about this last night.

I get a targeted visitor via G, due to my content G serves an appropriate ad, my visitor clicks on it and goes to the most crappy site ever and quickly leaves.

Under what circumstances would it be fair if I am held accountable for the owner of that crappy site not seeing conversions?

It seems under this concept the burden of making the sale is moved to the advertiser rather than the seller. If that's the case, I think advertisors should get more than 200 competitor bans.

If your site is offering the content where the visitor is looking for its better to use alternative ad (banners, textlinks)

Unfortunately, at least for the time being, my site isn't getting enough traffic to qualify for most of the a other programs.

Unfortunately, at least for the time being, my site isn't getting enough traffic to qualify for most of the a other programs.

you can show banners and text link from cj.com

[Join us @ facebook]

Finally this became a very interesting thread, and at this point I wish google marketers should take a look, as it seems that this convertion thing could create several problems.

I understand that G is trying whatever they can to prove to advertisors that they will ensure that publishers are not cheating the system. But there is no advertising system in the world that can promise 100% or any lesser amount of conversions; the only thing they could possibly do is to ensure x number of people are exposed (not read or even see - just exposed) to it. And to blame publishers who have no control over the ad text or the sales pitch given by the advertisors seems rediculous.

Think how that could prove play out. Why yes G, my ad campaign did bring me 1,000 visitors; but, none of them actually bought anything - so I don't think I should pay or at least I deserve a discount.

Hmm, think that would work with a newspaper, magazine or any other group that sells advertising space. Anyone think G will start SmartPricing themselves and give poorly performing campaigns a discount in proportion to the payouts to publishers?

[Join us @ facebook]

I think that Google has mixed things more than they should.

I mean that if you want convertions only, then you should be use a CPA program, not CPC.

Google propably wants to get in the CPA market, and they started that by adding to adsense their own products (Firefox, Picassa, etc.)

Propably the undergoing change is to let them get in the CPA market.

I think that Google has mixed things more than they should.

I mean that if you want convertions only, then you should be use a CPA program, not CPC.

Google propably wants to get in the CPA market, and they started that by adding to adsense their own products (Firefox, Picassa, etc.)

Propably the undergoing change is to let them get in the CPA market.

That is an interesting opinion, Nik.

Personally I believe that both advertisers and publishers are being observed by Google. If there are 4 clicks from your site to different advertiser sites and 3 of them meet conversion goals, then you should rank well with Google. Similarly for advertisers also they would be ranked based on performance. It should 'ideally' work but nothing is usually ideal!

[Guest]

[62.869]

Googlebot, hobbit, Yahoo crawler, Msnbot