I can’t tell you how many sites I’ve consulted on where, without a word to the SEM team, the site owner re-launches with a brand new site. Sometimes it’s a simple cosmetic change but more often than not its a completely new site with new page naming conventions and new folder hierarchy. In this article I am going to try and explain how you can minimize the impact to your search engine rankings when performing a redesign.

A typical scenario

One I’m sure most SEM’s are familiar with:

You receive a panicked phone call one day from a client. “My rankings are all gone, what have you done? My search engine traffic has dried up and I’m not getting any referrals!”

So the first thing I do is go to the site and guess what? It’s different than it was just a few days ago. So I ask the client: “When did you do the redesign?”

“2 days ago. Why?”

“That’s the problem. You redesigned without consulting me. You’ve essentially given the search engine a new website. That means the engine has to re-learn the site and how it works.”

Mind you I haven’t had to deal with this is a while, but I remember cases in the past where this has happened. So my first advice to someone planning such a redesign, no matter how minimal, is contact your local SEM or SEO professional and get their opinion.

Even a simple re-skinning of your site can have a huge impact on rankings.

How can even simple changes affect a site’s rankings?

Let me explain why:

First, let’s look at Google. Earlier this year, they released a patent that basically illustrates how much “smarter” Google will be. But in making itself smarter, Google is making it harder for site owners to make any changes.

Basically any change over a given threshold (and no, no one outside of Google knows what that threshold is) will trigger a review of the site, which means it gets removed, or lowered in the index. You may be OK with a 5% or 10% total change site wide, but anything over that will likely trigger such a review.

Because engines like Google can’t “see” the site, they have to compare the HTML to earlier cached versions to see if there’s been changes. When Google crawls a site (or page) they assign a calculated value, based on hundreds of elements, to the page. This is a fairly unique computational value, similar to a CRC value.

When Google comes back to request the page again, they perform this same calculation on the page and compare the CRC values. If they are different, they know there’s been a change and they need to revisit and re-cache the page. However, if there’s been a significant enough change (again breaching that predefined threshold), then Google needs to take a closer look at the page or site.

Yahoo! and MSN work similarly, however MSN tends to react a little quicker than Google. When MSN notices a change, the page(s) drop for a few weeks until the new changes are “assimilated” by the engine.

Page 1 of 3 :: First | Last :: Prev | 1 2 3 | Next