Google Website Optimizer 59
compuglot writes "Google has released the third leg of the stool in its quest to dominate online marketing. Google Website Optimizer is a multivariate testing application that allows users to test elements and combinations of elements in a website or landing page. The goal is increased conversions, and of course AdWords market share."
Re:been there? (Score:3, Interesting)
Re:Come on /. you can do better (Score:5, Interesting)
Especially Google's form of it. Slashdotters are coders and often want to put up interesting web sites to highlight their ideas, but that costs money, especially if significant bandwidth is involved. You can collect it from donations or support it with ads, and there aren't a whole lot of other options for sites which are interesting but don't have an obvious revenue model. Especially at the small scale, where the work of handling the revenue stream can distract you from doing the actual content of your site.
At least the AdWords are relatively unobtrusive, and targeted, which means that they may actually be of some interest to the people reading your web site.
Advertising is not evil. Flashing/spinning/dancing/up-popping/distracting advertising is evil. Polite, relevant advertising can be a way to support something without an immense amount of additional work.
Re:Come on /. you can do better (Score:4, Interesting)
Actually developers (at least those who do free software) have it very easy to advertise their work on Google: it's very simple and quite work-free to get a dingy little project page ranked very well on Google. All you have to do is list it on Freshmeat with the proper words in the project description and wait a couple of days. The huge number of sites that link to Freshmeat and archive the FM frontpage will automatically make a kajillion link to the project's page. I myself maintain a dozen small OSS projects that are almost invariably ranked very well in the Google first page when searching with fairly generic terms relevant to my applications.
So no, coders and nerds in general (the admitted target audience of Slashdot) don't need Google's marketing tools and don't need to pay a cent for them, because they benefit from the huge F/OSS social network on the net. Those who do need Google's marketing tools are those who try to *sell* you something that, unless the product is exceptionally good, isn't going to be listed at the top of the Google search unless the pusher pays Google.
That's why I say again that I (and I think most Slashdot readers) don't need/want to read about marketing tools.
Multivariate testing finally getting it's due... (Score:5, Interesting)
Understanding that Option A may work better than Option B isn't *nearly* as powerful as understanding that if you'd just taken certain components from them both, you'd have something even better still. Instead most marketers end up doing this endless Option A vs Option B stuff and never end up with what's really the "right" answer.
Then, there's the whole patience factor... most marketers don't have the simple willpower to put a test out there and let it run its course--especially when you've got so many options to test to do it right. Often, shortcuts get pulled because one particular version didn't work well, so it's assumed that derivative pages will also perform sub-par. (The reality is often surprising.)
Lastly, while we're on the topic of multivariate testing to my knowledge the only firm that has done proper, fully automated multivariate testing [memetrics.com] is Memetrics. Having worked with the so-called MVT solutions of other companies (which were mostly a joke) and Memetrics, too, Memetrics is the hands-down winner.
Google may have broader reach and even better marketing, but Memetrics is really a cut above IMHO.
Re:Multivariate testing finally getting it's due.. (Score:3, Interesting)
Other optimizations? (Score:3, Interesting)
Re:Come on /. you can do better (Score:3, Interesting)
So it's interesting not because Slashdotters will use this tool but because it's a useful development in the the way the web sites are supported. If the tool is useful, it's more advertising dollars shifted towards Google AdWords and presumably away from TV advertising (supporting useless network crud) or spinning/flashing/up-popping banners (supporting the kinds of web sites that tolerate that crud), but potentially supporting actual Slashdot users.
The fact that it's an interesting way to advertise web sites is also potentially of interest to Slashdotters, not for their own sites but for the sites they're paid to maintain, for those who do such things. Recommending marketing strategies may end up being part of their job. It's not really a technical thing, but we often wear many hats, especially in small companies, and as a web thing it often falls to us. There, too, advertising is useful; people only go searching for you if they already know you're out there somewhere.
Landing pages (Score:5, Interesting)
When that happens, not only do I leave the site immediately, resulting in wasted advertising money, I also lose faith in the overall relevancy of Google ads, making me less likely to click on any ads in the future. Generic landing pages aren't just a problem for individual advertisers; they hurt Google too.