Matt Cutts SEO Q&A 4

Matt Cutts – Heads the Google’s WebSpam team and works on Search Quality. Matt does a very nice Q&A session at GoogleWebmasterHelp’s Channel and would be very helpful for all of us on eBusiness.lk

Ok.. all eBusiness.lk SEO friends,

bring your pen and paper .. Class starts Now!

Lets closely listen to him and understand how Google works on Search Rankings.

Here are some of my(Niro) Notes to share with you,

Question :

Does the new canonicalization tag make it safe to add tracking arguments to some my internal links without fear that Google will split the quality signals between the two addresses? (Nick from Chicago)

• Matt believes that we can do it. But looks bit not too sure (I just felt that). Anyhow he asks to put this in a small section (eg:directory) and try. So everybody knows its completely safe.
• However if we can use some other way to track (eg:cookiee, analytics) – Its always better. Eg: Somebody might copy and link to it from their blog, but that tracking code which was included in the URL might not be valid anymore.
• Yes – he says we can safely use this tag (rel=”canonical”) as its what it meant for. So, if you have two pages that have same context but differ slightly only from the tracking code (eg, newmoon.com?code=SEMG and newmoon.com?code=SEMG&track=1234)
  

  use ,

  <link rel="canonical" href="http://www.newmoon.com?code=SEMG"/>

  That’s it!

Well don’t worry about this if you have only few pages on your site. I feel this is some problem we all encounter more often with big e-commerce web sites.

Down Side of Multiple URLs (because of using tracking URLs)

An e-commerce site might allow various sort orders for a page (by lowest price, highest rated…), the marketing department might want tracking codes added to URLs for analytics. You could end up with 100 pages, but 10 URLs for each page. Suddenly search engines have to sort through 1,000 URLs.

Then what happens ?

• Less of the site may get crawled. Search engine crawlers use a limited amount of bandwidth on each site (based on numerous factors). Many pages will be ignored and not get indexed because of this.

• Each page may not get full link credit. If a page has 10 URLs that point to it, then other sites can link to it 10 different ways. One link to each URL dilutes the value  the page could have if all 10 links pointed to a single URL.

Canonical URL best practices

We have to know that this gives search engines a hint, they don’t use this as a exact directive, (Google calls it a “suggestion that we honor strongly”) but are more likely to use  it if the URLs use best practices, such as:

• The  content rendered for each URL is very similar or exact
• The canonical URL is the shortest version
• The URL uses easy to understand parameter patterns (such as using ? and %)

Don’t abuse This

I know creative SEOs will think how to abuse this. How about getting less important URL to give similarity for the more important, using rel=”cronical”. Don’t waste time on this. First of all Google  reserves the right to take action on sites that are using the tag to manipulate search engines and violate search engine guidelines.And also the tag is taken to consideration only if the search engines also can give some assurance that pages match.