woaah.. looks complicated.

Grab a beer everybody - before you go through this stuff.

I havent seen it yet , put it seems like template stuff , you have just define some regions in the website and get a particular code from adwords , and make changes each time you make change in landing page in website optimizer only , without changing the actual code in landing page

not sure though , waiting for beta invite

So now Google condones cloaking as long as you're an AdWords advertiser.

Interesting.

I have found it to be interesting to implement but useful none the less.

Good for a-b testing on small traffic sites or button testing ie you have three button designs and or three locations to put them in and want to know which one converts better.

It is harder to implement on database + template backed content but possible if you take your time and read the directions and then the tips. You'll find your multi domain and multiple "success pages" are supported.

The key here is patience, or having the traffic to get really good numbers, if you get 800 visitors a day your better doing the A/B testing and give the tests a week or so to really solidify...

Nice addition and I'm sure this will come in handy, low traffic or not. It saves a bunch of work and I think alot of people who didn't test different versions in the past will definitely try this out.

I don't like the idea using google software to test conversions and show them the ROI. It will only give them a better insight of your market and your profit so they can raise the CPC for your specific market. Get own trackingsoftware .. its all over for cheap and better then this.

I don't like the idea using google software to test conversions and show them the ROI. It will only give them a better insight of your market and your profit so they can raise the CPC for your specific market. Get own trackingsoftware .. its all over for cheap and better then this.

Google already has quite a bit of info on each market segment and should be able to make pretty good determinations already based on users and historical trends.

But I don't have the system tracking sales so I am not concerned there and no matter the ppc provider my area has always been expensive, so once the competition starts to get tough - people start paying more per click when they are not number 1.

If there was no competition my clicks would still be 5 cents... And I would still track them using all the tools i have now as they each provide a different insight to help me close more sales.

how many people just tried googling for "timbuktu mountain bikes"
and got the fake adwords example :-)

We've used the optimizer with a database powered template system. So long as you put the tags in the right place you can pretty much control anything style related on the page.

We also have it set up to provide the same user experience across the board, although I'm not sure if this is supposed to be done.

Overall it provides interesting results, but to get any tangible results you either need very few combinations or a whole lot of traffic.

Also the conversion tracking is less then perfect.

Still for the price, and the interface you get, it's a good deal.

So is this served only to those who clicked on my adwords? Will organic traffic see this as well?
Last, what does googlebot see when it comes by, is it served my regular default content? Id be concerned making any drastic changes to pages that would also impact the organic positions. Sounds like a great tool though.

The tool is designed to test landing pages in order to increase conversion rates. Traffic could come from anywhere. That said, it is better to isolate traffic from different sources as much as possible to decrease the influence of traffic variance on results. Ideally, landing pages would exist independently of your main site. Landing pages have conversion as a goal and sales copy often works better than seo copy. Having seperate landing pages for marketing campaigns might be considered an online marketing best practice and also serves to seperate you organic optimization efforts from your conversion optmization efforts.

just tried your query AmznVibe...

got a Google server error 502 (Bad Gateway)

Does Google use this domain for testing variables..?