Chris Richardson of WebProNews attended this afternoon's Successful Site Architecture session, and sent some tips along exclusively to WPN readers.

Barbara Coll, CEO of Webmama.com and a former chair of SEMPO, led the session on Successful Site Architecture, a critical aspect of getting one's site prepped for search engine optimization.

Coll stated a goal of getting a site ranked highly in organic search, with the idea of converting visitors into customers. That SEO process has to be considered throughout the beginning of the site design, as well as adjusted later when the site is up and running.

Having developers on-board who can make you site sit up and be noticed helps, but Coll suggests that those versed in SEO development are ones a site owner wants to have on the project sitting in front of a workstation. The good practices implemented through design will carry through the life of a site.

SEO has to be embraced by the whole organization, from the CEO to the developers and content providers. It's an ongoing process, not just a destination.

Derrick Wheeler, Director of Search Optimization for Digital Impact, spoke next. He touched on the concept of knowing the little things about one's site, from the domains and sub-domains in use to the most important keywords and phrases in play.

Wheeler emphasized the importance of using robots.txt files to keep search engine spiders from hitting “spider traps” on sites that dynamically generate pages. Robots.txt can be used to keep spiders from wasting cycles on those areas of a site and direct them to your optimized content pages.

Session IDs, cookies, and javascript can hinder spidering. Robots.txt can be setup to turn off session IDs for spiders. And when redirecting visitors to parts of a site, use 301 redirects. Wheeler also reiterated the point that spiders like text links for navigation best; too much of a graphical focus on navigation may work fine for visitors, but spiders will get lost, and so will your site in the index.

Internal links should be short, as spiders may ignore overly long links, a potential sign of search engine spamming. File directories for the site should be somewhat shallow, three or four levels below root at most.

Peter Norvig from Google stepped up next, and the interest of everyone at the session perked considerably. As might be expected, Norvig urged site owners to utilize Google Sitemaps to help get their pages indexed quickly.

He emphasized some old school design concepts, ones that the Google spider considers when crawling: keep html pages under 100kb in size, eschew the use of splash pages, and don't use password protection like .htaccess for pages you want the spider to crawl.

Norvig reaffirmed Wheeler's contention on 301 redirect usage; use 301 when moving a page, and a 404 page after the page has been moved.

Eytan Seidman from MSN Search kept it simple. Build great content, and unique content. Get feedback from your users, and keep analyzing your site for issues. Seidman emphasized that MSN Search will remove sites from the index if they seem to be more about optimization that content.

Spamming search engines, link farming, keyword stuffing, content cloaking, hiding text, and duplicating content on multiple domains will all draw the ire of MSN Search, as well as other engines, and get a site deleted from an index.

Great info here. Thanks for sharing it. Makes one think of the things they have done and what they need to do. Thanks

__________________
YourWeb The Friendliest Shopping Mall On The Net |
Buying First Time | Whois Search

Please elaborate on this. If you have permanantly moved a page using a 301 why would you then go back and change that 301 to a 404? There could still be old links to the original URL and the purpose of the 301 is to point visitors to the new page.

Does this mean that after Google has picked up that a page has been permanantly moved we should turn off the 301 and replace it with a 404? Wouldn't that possibly result in a reduction if IBLs (if there are ones to the original URL) as well as visitors who don't get directed to the content they are looking for?

Good summary.

I think something got lost in the translation with this one. The robots.txt file can only be used to block spider access to pages or directories. I think the reference to turning off session IDs is related to the .htaccess file on Apache servers.

If someone knows how to do this with robots.txt, I think there are a lot of people who would like to know how to do it.

PHPfan makes a good point about 301s and 404s. You should be using a 404 when you want a search engine to completely remove a URL from its database and use a 301 when a page URL has changed. The 404 causes removal, while the 301 should cause the search engine to change the URL in its database.

__________________
Facts are meaningless. They can be used to prove anything. - Homer Simpson
MySQL Cheatsheet :: Arizona SEO training

You should have a 404 page setup regardless. So you either have the 301 or you don't.

A 301 permanant redirect will usually work okay for bots, but not so great for websites that may be linking to the page you moved. More than likely the webmaster will eventually find the link is broken and remove it.

I agree about having the 404 page. Every site should have one to capture users when a page is removed, but that's not what the statement appears to indicate.

That really doesn't have anything to do with "after a page has been moved". Like I said, every site should have a 404 page. If a 301 redirect is in place for a page that was moved, the 404 page will not do anything with respect to that page.

__________________
Facts are meaningless. They can be used to prove anything. - Homer Simpson
MySQL Cheatsheet :: Arizona SEO training