Another simple observation for discussion about how Google seems to operate at a very granular level using the same sort of scenario as yesterdays post.
- Q: Does Google Count A Keyword In The URI When Ranking A Page?
The anwser to wether a keyword in the URI makes a difference to whether a page ranks or not for a query, is YES.
And it’s it may be IMPORTANT from what I can see with other observations I have made.
Perhaps very important from a relevance point of view – though that is NOT WHAT MOST SEO SEEM TO THINK OF THIS – I didn’t until I looked for myself.
Think of it though – that makes sense. It’s the name of a whole document. It might be more important than the keyword on the page in plain text – that is, it might be more RELEVANT to Google. See my next screen shot.
- Observation – the keyword in the URI outranks a page with the keyword in the text content. Of course, this is just one test page, and positions can change with time due to any number of things. Why it’s important is another matter. For instance, does Google count this as a link – it IS cited on another document, albeit not in the anchor text of that link? Or is it just the keyword use itself?
Anyways – placing keywords in your URI is important as well all know, but the keywords in the URI – ie the filename of the document – don’t need to be on the page at all.
Does Yahoo count the keyword in the URI? Apparently so….
And so does Bing by the look of it……
Takeaways
- Google, Yahoo & Bing SEEMS to consistently rank a page in a search for a keyword that is only present in the file name (URL) and not present on the page itself or anywhere else, and that’s what I am seeing in a lot of places.
I’ve a few more I will publish. The aim of this is geek fun – it’s a search to find out how to make a page AS RELEVANT AS POSSIBLE so Google will rank it well, on a very basic level.
I already know how to make a page relevant, and I don’t sweat the small stuff. But I always like to keep an eye open for little hints like this observation MAY illustrate – because it is not JUST ALL ABOUT INCOMING BACKLINKS for a search engine demanding relevance.
Don’t just take my word for this or the recent meta description observation. See if this is true with your pages.
Of course – you don’t NEED to have a keyword in the URI. But search engines will count them if you do. Perhaps search engines place more emphasis on other regions of a page if keywords are not used in a URI. Who knows….?
Interesting that the 2nd result has the number string in the comments area, not the main article …
hi Shaun, Another great test! :) Since we’re talking about words in the url, do you think they need to be distinct (i.e. separated by commas)? I.E. bestdogfood is different than best-dog-food… Steve
Malcom I often wonder if Google treats the comment area as a secondary area. Steve, I usually use the ‘dash’ seperator as I think there is a big difference. I have a test that shows this – I will try and dig it out….. I lose more test pages than I remember lol
If you think about it… Google tries to serve up the BEST result for your query and if those two pages are the only pages in Google’s index to have your test phrase in it… then of course they’d rank – right? Google DOES look at the URL… but will a keyword in a URL outrank a competitive phrases that’s really been *gone after* wrt to SEO/SEM that’s NOT in the domain name? That would be a true test :) Just look at some of the domains out there that make no sense that have incredible ranking on some very competitive keyword phrases. Just thinkin’ out loud… Keli
You know Keli I am just thinking out loud. :) This is a seo theory blog – well it’s supposed to be. And of course, I’m just trying to poke into a few areas. It’s when you put all this (and a lot more) together – that’s when it’s of value.
Backs up a lot of the theory around niche marketing that targets single/ small groups of products in the domain name.
I totally agree! “It’s when you put all this (and a lot more) together – that’s when it’s of value.” Keep up the good work :)
Keywords in the URL are important, but their value has become lower than a few years ago…although keyword domains are still king.
While they might be important… I can think of sooooo many more techniques/practices that are much MORE important. That is so on the bottom of the list for me. As with math… you can eliminate the possibility of having a rule by providing an acception. I can do that in this case… totally whacky/strange domain names ranking extremely well for very competitive phrases. I have two *musts* when I advise a client on how to select a domain. 1) they should be easy remember… you should be able to rattle off your domain name to any *Joe* on the street and Joe should be able to to go home and reproduce (read: easy to spell) 2) no hyphens…. no one remembers hyphens In 14 yrs, it hasn’t failed me. Now if you’re trying to brand yourself with a product name, the above to tactics can be thrown out the window! Keli
Have you considered testing for the keyword saturation in the URL? Will it rank better for your search term if it’s taking up a higher percentage of the URL? For example, if I was writing a post that I wanted to rank for “blue widgets” Would http://www.testing.com/blue-widgets rank better than http://www.testing.com/get-your-blue-widgets-here
Document age? First discovered with term? Next stage of on page relevancy hasn’t kicked in yet? Link attribution hasn’t kicked in yet? Just a few concerns with what makes sense Looking to test CAPITALISATION and strong emphasis next?
On the subject of hyphens in domain names, I’m largely with Keli, but I’d go a bit further… Obviously, this is an SEO blog, so that is the focus here. But the usability aspect of hyphenating URIs shouldn’t be underestimated. I generally advise clients that they need both the hyphenated and the compound version of their chosen domain name. This way, they can give out the compound version over the phone (which requires fewer spoken syllables and results in less confusion e.g. Joe Blogs Enterprise Dot Co Dot UK). And they can put the hyphenated version on their literature (which makes it easier to read – compare joeblogsenterprise.co.uk with joe-blogs-enterprise.co.uk). Then just pick one to run the site on, and 301 the other one to that one. Best of both worlds.
Useful.
Thanks for posting this Shaun. I have been wondering about this for a while because I have been starting the practice of adding mini-sites via sub-domains to my main domains that have the targeted keywords separated by dashes as the sub-domain name. Sometimes tracking results in SEO is hard because of the fact that results are delayed. This is one more piece of evidence that makes me think that it is a smart thing to do. Also, I appreciate the example you added. That random number that you used really drove home the point.
I recently performed a site redesign for a client who in the past used not-so-great filenames. Choosing between renaming the files to something more user-friendly utilising keywords and potentially losing a small amount of link juice by 301ing them all (I think I remember Matt Cutts himself saying 9/10ths of link juice was preserved with a 301) or just sticking with the original filenames was a difficult choice. In the end I thought it would be best to keep the same URLs despite this being more complicated to implement because they were also moving from HTML files to .NET pages. I am wondering if I made the right choice… thoughts? Site hasn’t launched and all the URL re-writing work isn’t completed yet so it would still be possible to take a different course of action. The current site ranks reasonably well but it is a poor user experience and lacks proper branding, hence the redesign.
JORDON Do not go changing your URLS on the basis of this post without doing a bit of testing yourself. I did point out you don’t need to do the above to rank! Just consider it for the next time :) PS I hate the hyphen in my domain name – when I am telling folk what my email is.
I never doubted that this wasn’t the case. But it’s very handy that you’ve tested it just to prove it’s the case. Thanks!
Really great post Shaun. Yesterday’s post was great too. It is really great that you are sharing these great experiments with us!
Thanks for your reply Shaun. Don’t worry, I wasn’t going to make changes solely based on this! :) From the very beginning I wasn’t 100% sure of the best course of action, this just happened to be a timely post. The site already ranks and if I had any input in the site originally it wouldn’t have such shoddy URLs. Testing in this scenario could prove difficult though… Agree with hyphenation in the domain name, always more difficult to communicate. Email addresses based on names are bad enough e.g. someone could use Shaun, Sean, Shawn…
Thanks Shaun, this is something i’m going to experiment with.
A further benefit of hyphenation – avoiding potential embarrassment Just ask the guys & gals over at expertsexchange.com…
Yes hyphenated domains with keywords definitely can make it easier to rank for those keywords.
Great Post Shaun, I also have a hyphenated domain name, so I feel the pain regarding giving out your email and url over the phone. Even though nobody is suggesting that a keyword in a URL is going to throw any page off the top spot when they have a relevant link profile for that keyword, this is still handy info to know. Ps did you get the email I sent you about woorank? Bad timing there on my behalf as the tool seems to be having a few problems at the moment and is not working as it should/has far the last few months, bear with it though and check back, it looks like they are fixing it atm. Quote Mark: A further benefit of hyphenation – avoiding potential embarrassment Just ask the guys & gals over at expertsexchange.com… Lol thats pretty unfortunate, another good one is the italian power company…. powergenitalia.com
heh heh :)
[...] Does Google Count A Keyword In The URI (Filename) When Ranking A Page?, Hobo [...]
I think Google takes in to account every aspect before ranking. It surely includes this keywords too.
Well it seems to :)