You will be doubling your workload, essentially starting a brand new website from scratch in the same niche. I feel I need to make that very clear.

You asked about efficiency and that's pretty much the core of what I'm talking about, subdomains are inefficient.

Niche subdirectories are by far the better option.

It's a good idea but I would suggest that you really want that as part of your main domain rather than a subdomain. Any links pointing at it wouldn't be helping the main domain.

Now I don't know enough to suggest a fix, but a quick google found me this:

http://www.google.com/support/forum/p/Webmasters/thread?tid=5e5e79427977d86a&hl=en

I've never used mod_proxy, but apparently you can use it to have a different ip address for a subfolder. I believe it's ProxyPass Directive that you're after.

Hope that helps.

I can't speak for the Panda update, but I do agree with Daniel Deceuster, subdomains have been treat historically as separate sites. So unless things have changed dramatically you will essentially be starting from scratch link and work wise.

Subfolders may not be as neat or compartmentalised as sub-domains, but they are unambiguously under the ownership of the domain in question. Search engines can trust that.

Almost all of this work is about reducing ambiguity for the search engines, and sometimes that is at the cost of elegance.

Whilst I do agree with Daniel Deceuster, subfolders are historically much nicer than subdomains. You're not going to get in trouble for having one on a different IP address. We use a subdomain for our members section (decision was made before my time), customers like that mental partition, and it's not done us any harm.

I can see what you're getting at, but I agree with Lewis-SEO. Given that meta description is not visible on the page any changes (other than perhaps old school spammy) are just noted and displayed. If you were to change something on-page at the same time it may well get re-assessed, but that's not what we were discussing.

Even if it were to trigger a re-assessment all it would be doing is reviewing that page and putting it in it's "correct" place in the SERPs. Had you not done anything to it, it would still have dropped as soon as it was re-crawled regardless, all initiating a crawl would do is bring forward the inevitable.

Of course if we go on the theory that Google has started using user data; if the meta description changed how the page converted or how people reacted to the page (and you could check your analytics) it could effect your rankings.

Basically if users are now leaving your page in droves because of a misleading meta description that may have a knock on effect.

We've been running wordpress with the all-in-one-seo plugin up until now, but since looking at the Yoast plugin suggested here I've switched. Great suggestion.

Wordpress is very easy to get up and running, and can be very easily optimised, but if you want things to be "just so" it can be complicated getting it just how you want it. That said the volume of themes and plugins available make life very easy if you are willing to compromise. Just remember to keep it updated.

I am about to start playing with MODx which is really the other end of the spectrum, very customisable but with limited support compared to other CMS's. It looks clean and easy to optimise, but it lacks a lot of the bells and whistles that you can take for granted in other CMS's, simply because it's not as mature. They can probably be designed in, but you'll have to do the groundwork.

It really depends on what you're after.

You will be doubling your workload, essentially starting a brand new website from scratch in the same niche. I feel I need to make that very clear.

You asked about efficiency and that's pretty much the core of what I'm talking about, subdomains are inefficient.

Niche subdirectories are by far the better option.

I can see what you're getting at, but I agree with Lewis-SEO. Given that meta description is not visible on the page any changes (other than perhaps old school spammy) are just noted and displayed. If you were to change something on-page at the same time it may well get re-assessed, but that's not what we were discussing.

Even if it were to trigger a re-assessment all it would be doing is reviewing that page and putting it in it's "correct" place in the SERPs. Had you not done anything to it, it would still have dropped as soon as it was re-crawled regardless, all initiating a crawl would do is bring forward the inevitable.

Of course if we go on the theory that Google has started using user data; if the meta description changed how the page converted or how people reacted to the page (and you could check your analytics) it could effect your rankings.

Basically if users are now leaving your page in droves because of a misleading meta description that may have a knock on effect.

Whilst I do agree with Daniel Deceuster, subfolders are historically much nicer than subdomains. You're not going to get in trouble for having one on a different IP address. We use a subdomain for our members section (decision was made before my time), customers like that mental partition, and it's not done us any harm.