Andy Hagans and Partrick Gavin have a blog, Link Building Blog. Yes, I know, the title of the blog is vague and gives no indication whatsoever what the blog is actually about. It’s kind of like “Internet-Marketing-Blog.com” – very vague.
Anyway, Patrick Gavin is a friend. He runs Text-Link-Ads.com – you can buy text links, or sell text links on your own site. Patrick Gavin is a cool guy. Andy Hagans is a cool guy, too.
I left a comment:
v7ndotcomelursrebmem.tld is not an exact match. As far as Google is concerned, it might as well be chinesecheckers.com
That is, the words are not seperated by a hyphen so the keywords are not visible to Google.
Andy Hagans responded…
Something tells me that with 1000s of PhDs, Google may have figured that one out by now. I stand by my calling it an exact match.
And then Greg Boser stated as fact and without a doubt that Google does indeed parse keywords in URL’s and stated that they have been doing so for close to two years now.
Google has been identifing exact matches in domains without hyphens for close to two years now.
(Notice how the words in the domain are bolded even without a hyphen?)
Note the exasperated “please” in his response. Now when somebody states something with such confidence, I take notice, as do others. Obviously Greg Boser, placing a certain amount of value on his own credibility, would not make such a bold statement with proof.
Sadly, the only proof he did offer was to say, “Notice how the words in the domain are bolded even without a hyphen?”
This really doesn’t help, because it’s generally agreed upon that the highlighting is not indicative of algorithmic parsing of keywords. In order to prove that Google’s highlighting is non-algorthimic, all we need to do is find an instance of Google highlighting non-words. That is to say, insofar as parsing keywords requires Google to identify actual words as such, highlighting non-words will prove that the highlighting is simple highlighting and not parsing.
And here it is:
Webhos is not a word, but Google highlighted the occurence of those characters anyway.
Random letters, not a word, but Google highlights them anyway because Google’s highlighting is simply there to highlight. Not parse words – just highlight.
But all this is a side show. The question is not whether Google highlights or not – we know that they do. The question is whether or not Google parses keywords when they are run together in URL’s.
So here we go:
site:www.las-vegas-nv-online.com sexualfrustration – yields results on Google.
site:www.las-vegas-nv-online.com sexual frustration – no results.
site:www.las-vegas-nv-online.com sexual – no results.
site:www.las-vegas-nv-online.com frustration – no results.
Now the kicker –
las vegas nv online sexual frustration – MSN results. Obviously MSN is parsing keywords in the URL’s.
site:www.seattle-city.com paris hilton – no results
site:www.seattle-city.com parishiltonsexdogs – yields results.
But folks can say what they want about any of those search terms. Greg Boser posted in response to and in support of the claim that Google was parsing the keywords “v7ndotcom elursrebmem”, so why not test Google using those precise keywords?
Well, I did.
site:www.vw-lover.com v7ndotcomelursrebmem – results.
site:www.vw-lover.com v7ndotcom elursrebmem – no results.
In summary, Google does not parse keywords in URL’s when those keywords are run together. Pretty sure everybody knew this already, but there it is anyway.
PS. I also checked to see if Google was parsing keywords in URL’s when they are seperated by an underscore. The results were negative. With over 20 URL’s, each with keyword_keyword.html, and Google did not return them in a site search for the keywords. And some of these pages have been in Google’s index for over a year, so don’t give that “Google didn’t have time to parse the keywords” crap.