SMX Advanced 2011: The Really Complicated Technical SEO Infrastructure Issues

Moderator: Vanessa Fox, Contributing Editor, Search Engine Land

Q&A Moderator: Alex Bennert, In House SEO, Wall Street Journal


  • Jonathon Colman, Internet Marketing Manager, REI
  • Kavi Goel, Product Manager, Google
  • Steven Macbeth, Group Program Manager, Bing Search Quality, Microsoft
  • Todd Nemet, Director, Technical Projects, Nine By Blue
  • Maile Ohye, Senior Developer Programs Engineer, Google Inc.

This is one of my favorite sessions each year, primarily because Maile Ohye typically gives us some good easter eggs. And I like when Vanessa Fox laughs at dumb people, which she does in most other sessions as well. Anyways, let’s get started already!

Vanessa claims there is going to be a lot of “bonus things” in this session. #awesome

This panel is packed with awesome people. Now they are getting into talk. It’s fun to see Google and Bing people talking about These two dudes seem to be friends. Dogs and Cats! Living together! 40 years of darkness! Earthquakes! Volcanoes! …and that’s the end of my Ghostbusters bit. Anyways, looks like it might be big. Like MySpace big!

Now I’m watching a video. WTF is going on! This is awesome. Content in Context. So it appears that is a structured meta data foundation that will serve us potato salad.

Holy crap! talk has overtaken this session. The session is no long stuctured at all. It’s just a bunch of questions. Getting tough to tweet and blog. And now they are talking about the new author tag from Google that was announced today.

At any rate, here are some of the questions:

  • If they are moving to microdata? will effectively make microformats obsolete?
  • What do we do with the tags and standards? Do we go back and mark up our entire site?

Oh wait. The talk was only part of the session. So I guess now we’re back to the original session topic. Up first is Maile Ohye.

Maile Ohye is going to talk about Internation SEO issues:

  • Considerations for international expansion of your site
  • Before you start, ask yourself if you can really support that site.
  • Key factors:
  • Flow chart: Language or country/region
  • Language focused expansion:, put site on gTLD or ccTLD if you can afford it
  • Country focused expansion: language still remains a factor. Multiple languages in one country. Currency, shipping, local laws.
  • Can you get ccTLD for your site?
  • Check out United Airlines site. Maile mentioned that they “are doing it right” (I’d take her word and use their site as a model)
  • No cloaking! Lots of sites do this. Don’t.
  • Shareable URLs.
  • Indicated language/country in the URL (
  • Matching URL structure. (ie. for all countries)
  • Helpful language/country crosslinking
  • Provide user ability to navigate to their desired language of choice
  • Site load times factor in
  • Geo metatags not used by Google with web search? (Well, what are they used for? #confusing)
  • Duplicate content ice cream: on different ccTLDs, duplicate content shouldn’t be a problem

Next up: Jonathan Colman

  • “I’m like you. I’m also a non-developer.”
  • Jonathan hit a homerun with his rel=canonical tag idea, and he increased his pages indexed while eliminating a lot of duplicate content.
  • He saw massive increases in pages indexed, while also seeing a 96% decrease in duplicate content.
  • Now it’s Vanessa and Maile addressing what Jonathan talked about. Maile does not recommend what Jonathan does.
  • I’m really confused now. I feel bad for Jonathan. He’s a badass. And I hope his site doesn’t get blown up this week by Maile.

Next up: Todd Nemet

  • Really long redirect chains, unintentional referrer cloaking, IIS issues, robots.txt issues, IDS blocking Googlebot
  • Redirect chains:
  • Crawlers give up after too many redirects
  • AdWords adds even more redirects
  • Cloaking:
  • By referring URL
  • By user agent
  • By IP address
  • IIS browscap cloaking – look it up
  • IIS error page handling:
  • creates 302 chain out of the box for 404 pages

Then the Q&A was all about rel=canonical stuff. I’ve heard it a hundred times before. Google created a monster when it created the rel=canonical tag.