These days there is nary a site which is being redesigned or redeveloped without at least some measure of concern for its search engine friendliness and optimization. The equation is simple, you have content which there is likely free, relevant youb traffic for… and you want it. Even if you don’t want to pursue an aggressive strategy for building links and doing heavy on-page optimization, you should make sure to cover your bases so that at some point, should you wish to make search optimization a priority, you will have an appropriate foundation to do so.

In order to not make this the “seo site health bible for re-launching a site”, I’m going to focus on four pillars of a smooth transition which should preserve or improve your site’s search engine positioning, and set things up to stay in good search engine health, all other things considered.

Trying to Maintain the Previous Site’s URLs
This basically means that if the current site has decent, static (looking) URLs it is usually best to just stick with those, adding new URLs only according to what “brand new” pages are going to be added. Doing this means that you won’t have to 301 redirect as many pages (the search engine benefit of 301 redirects can take months to kick-in in some cases), and you will maintain any ranking benefit that may occur due to the age of a URL, etc. This also decreases pageload times as your server won’t be under stress from tons of redirects (like one recent client with 17,000 unique articles which had to be redirected).

If the current URLs are not looking pretty and especially if you deem them to be causing indexing problems, please see below: “Redirecting Old URLs…”.

Launching with Static URLs
In order to never let on to a search engine that your site is dynamic and what the dynamic locations (URLs) of our pages are, you need to be careful not to launch a site before creating the proper URL rewriting so that all pages use a static looking URL and also you must be sure that all internal linkage points to these static versions ONLY.

Once you get pages indexed the search engines can be very steadfast in holding onto them. Trust me, I know you told your client you’d launch the site tonight… but explain to them that there are a few last-minute SEO related issues that you must cleanup in order to maintain site health; finish getting all of the URL rewriting done, then launch.

No Duplicate Content
You must always insure that no two URLs show the same content. This often includes instances like http://www.example.com/index.html and http://www.example.com/ (the correct URL if you want the www. included.). This also includes instances like http://example.com/ which again, should be http://www.example.com/. Please note that if you try doing that with http://websandiego.org/ it will correctly redirect all versions to http://www.websandiego.org/.

So, the important steps for checking this issue off our list when developing/redeveloping a site are:

  1. Implementing /index.htm, /index.html, /index.php (or other default pages) 301 redirects to / (please note that this should occur if you use directory index pages as well like http://www.example.com/main-category/ [you do not want the same content to be accessible at /main-category/index.html])
  2. Implementing non-www. to www. 301 redirects (in some cases like http://performancing.com/ you would do the opposite since for a long period of time the non-www. version has been the preferred URL).
  3. Make sure that all internal linking reflects these preferences (e.g. a common mistake is that the sitewide homepage link points to /index.html or some equivalent of that. It should at least be / but more preferably should be an absolute URL like http://www.example.com/. So, as you can see, if you’re going to do absolute internal links you must also be sure to reflect the preference for either www. or non-www.
  4. Make sure there are no internal links referencing the dynamic version of any URL. You can do this by using something free like xenu link sleuth to crawl the site.
  5. Preferably make sure that the dynamic versions of URLs are not allowed to be accessed. This means that you should either 404 them or 301 redirect them to the correct version. This may not always be realistic with some CMSs and budgets that don’t support the effort, but this would ensure that there is no SEO sabotage done by competitors who maliciously link to these URLs to cause duplicate content penalties (or by any other means of someone accidentally linking to these URLs).

Redirecting Old URLs which are not Used on the New Site + Custom 404 Pages
This issue requires a few steps. Basically our goal here is to 1) conserve existing PageRank (link popularity) by avoiding losing (404ing) any pages which have external inbound links pointed towards them; 2) having visitors which do get 404 error (page not found) be presented with a page containing links and graphics, etc so they don’t simply hit back (assuming they came from a link, search engine, or anywhere externally).

**Note*** The method described below is actually describing what we have a bot built to do. So, the point of displaying is it is that you should interpret what we are doing and what we’re trying to accomplish and do as much of this by hand as possible (or as you can bear). If you have a large website hopefully the client’s budget supports building a script like what we have.

The steps are as follows:

  1. Crawl all the backlinks (up to 1,000, or get creative by adding in various terms to uncover more, then remove duplicates) pointing to the site you’re launching.
    • Then will crawl each result to determine which exact page, on the site you’re launching, the link is pointing to.
    • Then create a list of redirects (you will redirect these pages to their relevant counterparts or to a sitemap type page) for you to implement.
  2. Attempt to find all known URLs and Existing URLs
    • Spider the current site for a list of all URLs (with something like Xenu).
    • Do a Google or other search engine “site:” search and create a list.
    • Analyze and determine what, if anything to do with this URL list (possibly some specific redirects) .
  3. Create a custom 404 page.
    • This page will basically be a page which matches the site’s design, displays a simple message saying the page could not be found and may no longer exist. This page will have all the standard links so that visitors can quickly get where they want to go.

These are the pillars of starting off with good search engine health. Anything else can go into effect very quickly so long as this stuff is totally taken care of by launch, so don’t worry about getting it all done if there are crucial deadlines (that way us SEO guys aren’t getting blamed for everything).