Analytics & Testing, Search Marketing

WWW or No WWW and Pagespeed

For the past few months, I’ve been working to improve my site’s pageload time. I’m doing this to help improve overall user experience as well as to help my search engine optimization. I’ve written about some of the methods I’ve used for speeding up WordPress, but I’ve also changed hosting companies (to Mediatemple) and implemented Amazon’s S3 services for hosting my images. I also just installed WP Super Cache at the recommendation of friend, Adam Small.

It’s working. According to Google Webmasters, my pageload times have decreased to well within Google Webmaster’s recommendations:
www-pagespeed.png

Google also lets you set the default for whether or not your site is set to go directly to the www.domain or without the www. This is where things get interesting. If I observe my page load times without the www, they’re fantastic. However, if I look at the pageload times with the www, they’re terrible:
www-pagespeed.png

The irony of course, is that the hosting package I have always goes to a www page. Because of the huge difference in Google’s response times, I’ve set the site configuration to the non-www address within Google Webmasters. I also removed redirect code in the root of my site in the .htaccess file that was redirecting non-www requests to a www domain.

I’m not sure if any of this helps or hurts, but it seems the logical thing to do. Any thoughts?

8 Comments

  1. 1

    This is very interesting! I always redirect my websites to the WWW version for consistency and to give Google a single URL to index so rankings are not split. I also think it looks better and more balanced to the eye to force the WWW version to display. Your data, however, makes a compelling argument to re-think this. I would be curious to see your SEO results after some time. I would love it if you would share them here after some testing.

  2. 2

    Odd…just now I was reading another post and wondering why the page was taking so long to load. Looks like cdn.js-kit something was taking forever. According to your graphs, ooks like whatever you did is helping!

  3. 3

    That’s my commenting package, Joshua! I’ve seen some lag with their service as well and may have to say something soon.

  4. 4

    Will be glad to share any stats Michael! Once again, though, everyone IS going to the “www” address so I’m not sure why Google bots are slow to get access that route. Wondering if its a nameserver issue with my hosting or an apache setting or something.

  5. 5

    Yahoo! recommends using WWW. to allow for non-www. static image domains:

    <blockquote cite=”http://developer.yahoo.com/performance/rules.html#cookie_free”>If your domain is http://www.example.org, you can host your static components on static.example.org. However, if you’ve already set cookies on the top-level domain example.org as opposed to http://www.example.org, then all the requests to static.example.org will include those cookies. In this case, you can buy a whole new domain, host your static components there, and keep this domain cookie-free. Yahoo! uses yimg.com, YouTube uses ytimg.com, Amazon uses images-amazon.com and so on.</blockquote>

    Ever since reading this, I’ve gone with http://www….because Yahoo! is pretty smart.

    This is the first I’ve heard of any www speed issues. Anyone else have the same experience?

  6. 6
  7. 7

    I force with the no “WWW” so my domain is simply my name. I haven’t really tested it for speed reasons, but anytime you visit my site you get the no “WWW.”

    I looked at it from a branding perspective. I think for businesses – the “WWW” puts a perception of dependability.

    I’m half-tempted to test for speed myself. I have noticed my site loads pretty quickly on a regular basis. Coincidence?

  8. 8

Leave a Reply