Jump to content

  • Log in with Facebook Log in with Twitter Log In with Google      Sign In   
  • Create Account

Subscribe to HRA Now!

 



Are you a Google Analytics enthusiast?

Share and download Custom Google Analytics Reports, dashboards and advanced segments--for FREE! 

 



 

 www.CustomReportSharing.com 

From the folks who brought you High Rankings!



Photo
- - - - -

Unique Urls


  • This topic is locked This topic is locked
86 replies to this topic

#31 webd

webd

    HR 2

  • Active Members
  • PipPip
  • 46 posts

Posted 25 December 2007 - 11:48 PM

QUOTE(Jill @ Dec 25 2007, 02:05 PM) View Post
Clearly that search does show that there is some weighting placed on the information in URLs. Which makes good sense. The question we generally have is exactly how much weighting?


Like I mentioned, it is not worth changing existing URLs, but if starting from scratch or rewriting the URLs for other reasons, I think it is definitely worth adding keywords in the URL.

I think that keywords in URLs is more important than the meta keywords tag because Google provides more traffic than Yahoo, and I think that more weight is given to keywords in URLs by Google than to meta keywords tag by Yahoo. That last part is harder to prove though. Makes sense to use all the tools in the toolkit.

#32 Randy

Randy

    Convert Me!

  • Moderator
  • 17,540 posts

Posted 26 December 2007 - 09:50 AM

QUOTE
Makes sense to use all the tools in the toolkit.


Up to a point, I'd agree.

When it's a new site and/or you have a blank canvas to work with because there's going to be a complete site revamp I'd agree fully.

However from what you've said elsewhere I'm guessing you'd agree that it wouldn't be a good thing to destroy already good rankings for an established site just to get keywords in the urls. The rule of diminishing returns kicks in pretty quickly for these situations.

#33 webd

webd

    HR 2

  • Active Members
  • PipPip
  • 46 posts

Posted 26 December 2007 - 10:38 AM

QUOTE(Randy @ Dec 26 2007, 10:50 AM) View Post
When it's a new site and/or you have a blank canvas to work with because there's going to be a complete site revamp I'd agree fully.

However from what you've said elsewhere I'm guessing you'd agree that it wouldn't be a good thing to destroy already good rankings for an established site just to get keywords in the urls. The rule of diminishing returns kicks in pretty quickly for these situations.


Yeah... we agree there... It would be a bad idea to change URLs on a site just to add keywords to the URLs.

#34 mcanerin

mcanerin

    HR 7

  • Active Members
  • PipPipPipPipPipPipPip
  • 2,242 posts
  • Location:Calgary, Alberta, Canada

Posted 15 January 2008 - 03:08 AM

BBCoach, your test seems to show that a keyword in your URL harms rankings, which quite honestly makes me question the methodology, since it conflicts with both my own real-word experiences and what I believe to be common sense. I think that there were other factors involved.

If keywords in URLs did not matter, then the results of an accurate A/B test should be that there was no difference in rankings at all. This would support your "pointer" theory.

If keywords in URLs had a positive affect, then the keyworded URL's should show better rankings.

You say that none of the new URL's are ranked. If this is an accurate test, then that proves not that keywords in URLs are a pointer, as you claim, but in fact are a significant factor in rankings, and indeed clearly result in negative rankings. Your conclusion is at odds with your own data.

Not to worry, because I honestly suspect the data. From a common sense viewpoint, it beggars belief that the keyword metatag does nothing, but keywords in URL's actually harm your site. It simply does not make any sense. It;s not personal in any way, but simply an honest assessment of the experiment as described.

From a real-world point of view, I also have a ton of data from sites that include 2 governments, the third largest online pharmacy in the world, and many others. Absolutely none of it supports negative effects on rankings due to keywords in URLs. All of it supports at least an inclusionary effect.

As for a rankings effect, the result is either neutral or positive, but appears to be slightly positive. The competitiveness of the sites involved and the volume they do gives me very fast feedback on even small changes, so I am pretty sure of this, and it's consistent across several industries. I will also say that there is no evidence of a magical huge gain in rankings, either, based on the same data.

As others have mentioned, I'd use them for a clean-slate site, but I'd think long and hard about redoing all the URL's in an already well-indexed and successful site - the effort and time is best applied to getting links instead.

Ian

#35 worlddom

worlddom

    HR 1

  • Members
  • Pip
  • 6 posts

Posted 26 January 2008 - 06:35 AM

If I'm looking for a kenwood mixer then 90% of the primary results have 'kenwood mixer' within the url structure.

Not only that, from a none SEO perspective, the keyword is highlighted... which has to improve ctr

#36 BBCoach

BBCoach

    HR 5

  • Moderator
  • 402 posts

Posted 26 January 2008 - 03:44 PM

Sorry Ian for not responding sooner. I haven't paid attention to this discussion since my last post.

QUOTE
You say that none of the new URL's are ranked. If this is an accurate test, then that proves not that keywords in URLs are a pointer, as you claim, but in fact are a significant factor in rankings, and indeed clearly result in negative rankings. Your conclusion is at odds with your own data.

Not to worry, because I honestly suspect the data. From a common sense viewpoint, it beggars belief that the keyword metatag does nothing, but keywords in URL's actually harm your site. It simply does not make any sense. It;s not personal in any way, but simply an honest assessment of the experiment as described.


It's worse than not being ranked. I can't find them in the index at all. Personally, I think G/Y/M ignores them completely since the rest of the URLs on the site are not built that way. But the fact remains those URLs are not in the index after 12 months. Therefore, it's not at odds with my data or with my statement that URLs are pointers to information. Bottom-line for me is if it works then I'd do it. I'm not into arguing my beliefs only what I understand and know. If I see there's a significant advantage in using keywords in URLs, then I'll jump on it like a frog on junebugs. However, I've also built two other sites with 2-3 keywords in the URLs with no proof there's an advantage to those pages rankings in the SEs after two years of being live. Yeah those two web sites are ranked, but they're not #1 (sometimes not in the top 10) if you searched for those URL keywords and some of those pages should be. From my tests and observations it doesn't have enough of an impact for me to build ecommerce sites that way or make a recommendation to someone to do it. To me it's just not that beneficial if all of the other SEO strategies are followed.

BTW, running any tests against government web sites I suspect would not yield "honest" results because all government web sites are THE SOURCE for that government which serves millions of people interests and needs. They automatically get special treatment from the SEs as a result of that (i.e. their own groups) because SEs feel that it's a valuble service they offer, not to mention zillions of links pointing back to their information. Wished I had one government web site in my pocket. Now the pharmacy one is something else. Would you be willing to expand/quantify your statement, "All of it supports at least an inclusionary effect." for the pharmacy site? How much of an inclusionary effect is there? Have you tried building a page on it that didn't have keywords in the URL to see how it would be treated in the SEs versus the ones with keywords?

Also, don't worry about saying "it ain't personal." From what I've observed you're very professional and well spoken with significant experience and testing to support your positions. I always read and learn from your posts. Especially the international stuff where I have no experience.

Edited by BBCoach, 26 January 2008 - 03:50 PM.


#37 Jill

Jill

    Recovering SEO

  • Admin
  • 33,244 posts

Posted 26 January 2008 - 04:03 PM

QUOTE
BTW, running any tests against government web sites I suspect would not yield "honest" results because all government web sites are THE SOURCE for that government which serves millions of people interests and needs. They automatically get special treatment from the SEs as a result of that (i.e. their own groups) because SEs feel that it's a valuble service they offer, not to mention zillions of links pointing back to their information.


I'd surely like to see some proof of that bold statement BBC!

#38 BBCoach

BBCoach

    HR 5

  • Moderator
  • 402 posts

Posted 26 January 2008 - 06:22 PM

Well Jill go to G/Y/M and checkit out. They have their own groupings or sub group searching for government web sites created by the SEs manually, and they do have zillions of web sites linking back to their information. I think it's pretty obvious. I reckon I could be wrong, but I don't think so. Do a query for index.html and see who's in the top 10. I don't think that it is that outrageous of a statement. Test it. Type any state name and state web sites ie [new hampshire state web sites]. Looking for all of the web sites in a state. Who's at the top? Type [new hampshire state taxes]. Who's at the top? You can pretty much go through all fifty states with the same results. Same with the US government, or the British, French, Russian, etc. My opinion is that government web sites have a huge advantage in the rankings versus anyone else, but like I said I could be wrong and it's just an opinion. BTW, if you find one that doesn't, let me know and I'll contact them for SEO consultation. biggrin.gif

Edited by BBCoach, 26 January 2008 - 06:28 PM.


#39 Randy

Randy

    Convert Me!

  • Moderator
  • 17,540 posts

Posted 26 January 2008 - 07:53 PM

I suspect the idea Jill may be having an issue with is if a .gov TLD carries any extra weight. But I'm not sure BBC is saying that a .gov domain would do any better than a .com with the same positives being part of the picture either.

Government sites do have a lot of ancillary stuff going for them, including lots of high quality links from lots of places. Quite honestly they don't need any boost from their TLD, and I'm pretty sure you couldn't get everything down to that one factor for testing even if you tried really hard.

#40 Jill

Jill

    Recovering SEO

  • Admin
  • 33,244 posts

Posted 26 January 2008 - 10:30 PM

Randy, I agree that government sites would naturally do well for a variety of factors. My beef is in the assertion that the search engines are actually giving them some sort of special boost beyond typical algorithmic means.

I would be highly surprised if they were, but as I haven't tested it myself, I'm certainly interested in BBC's findings if he has indeed tested it. (Or if anyone else has.)

#41 BBCoach

BBCoach

    HR 5

  • Moderator
  • 402 posts

Posted 27 January 2008 - 12:05 AM

QUOTE
My beef is in the assertion that the search engines are actually giving them some sort of special boost beyond typical algorithmic means.
Ok. I can understand that. BUT, why do SEs provide a SUBSET of a search for government sites if they don't have an advantage? I think they do and it would be logical to do so as a SERVICE to the public. As far as this subject goes I don't think this is an issue. I was merely pointing out that gubment web sites would not be as reliable of a test as non-gubment sites because of their special status. If you don't think so Jill then fine. I just think the dice are stacked against other sites that would echo gubment sites because of the relevance of THE SOURCE. Particularly since most of my hobby pages compete against .gov sites. It's an observation. Not a rule. And yes I've tested it against my hobby site. Gubment sites are hard to beat and I compete against them directly with my hobby site. It's an opinion (with only observations). Test it and see who's top dog for the different governments.

#42 BBCoach

BBCoach

    HR 5

  • Moderator
  • 402 posts

Posted 27 January 2008 - 01:38 PM

I forgot to say that I don't believe the .gov TLD has any more juice than an other TLD. Some of the government web sites that I compete against have .us, .com, & .org TLDs. I dunno why they all didn't use .gov, but not all of them do. Mostly they use [department.state.ma.us].

#43 Jill

Jill

    Recovering SEO

  • Admin
  • 33,244 posts

Posted 27 January 2008 - 03:30 PM

Personally, I just think that the official government sites are going to naturally have the best, most authoritative links, which is why they do well.

#44 BBCoach

BBCoach

    HR 5

  • Moderator
  • 402 posts

Posted 28 January 2008 - 12:16 PM

QUOTE
Personally, I just think that the official government sites are going to naturally have the best, most authoritative links, which is why they do well.
Jill who's more "authoritative" than a gubment web site? They may have a zillion links from lawyers, accountants, news, and lobby/hobby web sites, but those are not more authoritative than the gubment site itself. I do agree that "naturally" gubment sites will kick some booty, but I also think because SEs have manually created sub category searches for gubment web sites indicates a partiality to THE SOURCE.

Question. Do you think a test conducted against one of those powerhouse web sites prove the test conclusively? I don't think so, because even if they're not given a V8 boost by SEs, they're naturally getting it as we both agree. A test on whether or not keyworded URLs help with ranking would be to do it on a less popular site than a gubment site. That's what I was implying by that statement.

#45 Jill

Jill

    Recovering SEO

  • Admin
  • 33,244 posts

Posted 29 January 2008 - 12:03 AM

QUOTE
but I also think because SEs have manually created sub category searches for gubment web sites indicates a partiality to THE SOURCE.


Then we'll have to agree to disagree on this one, BBC!

I do believe Google when they say they don't provide any special treatment for any site.




0 user(s) are reading this topic

0 members, 0 guests, 0 anonymous users

We are now a read-only forum.
 
No new posts or registrations allowed.