When it comes to the search engines and country-specific content or searches, there are several obvious variables the engines use to determine the geography of a particular site or company:

  • TLD (.com or .co.uk or .ca for example)
  • Hosting (Is the site hosted in the UK, the US, Canada?)
  • Inbound links (Are they primarily from US sites, UK sites, Canadian sites?)
  • Spelling and grammar (Do they use the metric system? Is color spelled with a "u"? Do they put "eh?" at the end of sentences? - kidding, Canadians, only kidding.)
  • Language (if specific to a country/region, which excludes English, French or Spanish)
  • On-site addresses, phone numbers (which only works if your site doesn't list lots of International addresses as well)
  • User base (does the majority of traffic to the site come from a particular geography?)
  • Address verification (if the company opts to be listed in Google local or similar)

There are probably several others as well, but, as 2K has pointed out in the past, search engines are not overly responsive to even the strongest of these input criteria and appear easily confused about what country or region a site is actually targeting.

So, let's work from a hypothetical... Assuming you had an English-language site that had content for Canada, the US, Australia and the UK, would you opt for:

  • A: One TLD (.com) serving all the different content on subfolders (i.e. site.com/ca, site.com/au, etc.)
  • B: 4 unique TLDs, all serving their unique content (i.e. site.au, site.co.uk, site.ca, etc.)
  • C: Another option...

In scenario A, you lose the geo-targeting, but gain the value of all links providing authority to a single site. In scenario B, you spread out authority and link value, but have successfully geo-targeted and maintain that advantage should the search engines ever get more serious about regional results.

p.s. Sorry for the long delay on posting; Matt and I have been galavanting in Vancouver :)