SEO Tycoon

**juliensimon** · 07-24-2008, 08:00 PM

Let's say I set up an A/B experiment set-up on the home page. What are the effects on the SEO regarding indexing pages, ranking etc...

**bill benson** · 07-24-2008, 08:14 PM

How are you going to deal with the duplicate content aspect in an AB experiment?

**VirtuosiMedia** · 07-24-2008, 08:29 PM

What are you testing for? In general, I think that A/B testing is done for testing conversion rates for traffic and not for SEO. I wrote an article on my blog a little while ago about why generalized A/B SEO testing is next to impossible because of all the factors that are outside of your control.

**chrishirst** · 07-25-2008, 04:10 AM

And a "home" page is not the best place to be running split testing on.

**juliensimon** · 07-25-2008, 11:35 AM

Quote:

Originally Posted by bill benson

How are you going to deal with the duplicate content aspect in an AB experiment?

Well, the point of an A/B experiment is testing different content etc...

So I'm not too worried about duplicate content.

**juliensimon** · 07-25-2008, 11:38 AM

Quote:

Originally Posted by VirtuosiMedia

What are you testing for? In general, I think that A/B testing is done for testing conversion rates for traffic and not for SEO. I wrote an article on my blog a little while ago about why generalized A/B SEO testing is next to impossible because of all the factors that are outside of your control.

I am not testing it for SEO but traffic and conversion rate as you said. One of the website I am working on has a very high bouncing rate and low conversation rate. I am redesigning it, changing the whole navigation process which is a mess right now. I wanted to know if it's worth spending time testing different pages for the new pages.
But I am wondering how it will affect the rankings of those pages

**juliensimon** · 07-25-2008, 11:41 AM

Quote:

Originally Posted by chrishirst

And a "home" page is not the best place to be running split testing on.

That's what I thought to. But right now, there is a very high bouncing rate which mean the main landing page (usually the home page on this website) is not attractive to users.
Having said that, that home page actually ranks pretty well, drives a lot of traffic but doesn't "keep" it. So I don't want to mess around with it too much and make it drop in the SERPs.

**chrishirst** · 07-26-2008, 05:18 AM

Quote:

here is a very high bouncing rate which mean the main landing page (usually the home page on this website) is not attractive to users.

Not necessarily, a high bounce rate is pretty average for a "home" page. Split testing the "home" page will probably mean you half the "bounces" on each version but still maintain the same percentage.

If the site is one of the ones in your signature, check the referrers on the "bounce" hits. Odds are there will be a lot from here (and other forums you may post at).

Now IF the bounced visits come from search engine queries with a related search word/phrase THEN you could be seeing a potential issue.

**juliensimon** · 01:17 PM

Quote:

Originally Posted by chrishirst

Now IF the bounced visits come from search engine queries with a related search word/phrase THEN you could be seeing a potential issue.

I checked the analytic stats and the bounce rate for visits coming through search engines with related keywords is around 70%.So yeah, there is a problem, but I am working on it. I am redoing the website completely because the navigation is a big mess. Also some content needs to be updated etc etc...