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

#16 BBCoach

BBCoach

    HR 5

  • Moderator
  • 402 posts

Posted 03 November 2007 - 01:05 AM

This URL "http://www.myname.co...tie shoes.html" is not any more valid than "http://www.myname.co...om/123456.html"

They are both pointers to some kind of shoe tying information (hopefully). Who's information is more relevant to a given query is based on the resulting page and the SE's algo and not the URL to the information. Many a person has been led down this path with great disappointment thinking their URL will help with rankings. It's the CATEGORIZATION (DEWEY DECIMAL SYSTEM) LANDING PAGE that counts and not the URL to that page.

BTW, don't think for one second I'm meaning that LITERALLY you should use the Dewey Decimal System to organize your web site. However, every industry can be categorized from major to sub categories for taxonomic reasons. Feed the SEs hierarchical and easy to digest meals and the humans behind the scenes will confirm the relevance. At least according to the patents and my log files.

Edited by BBCoach, 03 November 2007 - 01:15 AM.


#17 Jill

Jill

    Recovering SEO

  • Admin
  • 33,244 posts

Posted 03 November 2007 - 08:10 AM

BBCoach, did you link to the keyworded URLs in the exact same manner as the non keyworded ones? In other words, did you perhaps make half your product URLs with keywords and half without?

If so, it sounds like a good test.

My concern is that if you created the keyword rich URLs as extra ones, on top of the already existing non-keyworded ones, which would cause it's own set of problems.

My hope is you did the former, and if so yahoo.gif because I've always felt that was true, but had never done any kind of extensive test like that.

#18 BBCoach

BBCoach

    HR 5

  • Moderator
  • 402 posts

Posted 03 November 2007 - 12:56 PM

Jill,

Yes that's the way we did it. The compromise between marketing and my purple fit was that it wouldn't effect the entire web site, but only NEW categories. Ten months ago we added one new parent category with up to 4 levels of sub-categories in the same exact structure as the original categories. The only difference is the product links are now like this: one+two+three+four+five+six+seven+eight+nine+ten+eleven+plusabunchmore.html. The products are NOT indexed and ranked based on the description of the product URLs. They're indexed and actually poorly ranked based on the product page information. I think the top 3 SEs may even possibly ding these links because they interpret those links as too much SEO. My opinion with no proof and my Gneer friend won't confirm or deny my opinion.

Why do I have that opinion? Because new products in the original categories are indexed and highly ranked within a few days of going live on the site, and those NEWLY hatched URL keyword stuffed (with product titles and categories) products are not even on the radar screen for weeks after going live. Trust me. I'm working real hard to prove that putting keywords in a URL is not a benefit and it just might actually be a negative. I have plenty of log file data and am manually parsing it out in a very granular fashion to show the relationship between the SE crawls, the SE indexing/ranking of products and the SE referrals to products. What makes it easy is that each of the top 3 SEs crawl all product pages on a monthly basis. So I have month to month comparisons to work with and will gleefully shove my reports up their noses when 12 months of this stupid+no+benefit+and+A+LOT+of+work+scheme+is+done. Much, much more important than a URL is the taxonomy of a web site.

I think this keyword URL stuff came about because of the way Ebay and other auction sites create their URLs and when the marketing geniuses saw the high rankings of their links in SEs they assumed it was because of their links. Never mind the fact that auction sites are for the most part taxonomically spot on and that Ebay and Google created a marketing partnership. Well at least until Google tried to smooze Ebay's PayPal partners away. When will people quit trying to get over on the SEs and do the work necessary for good rankings? All three top SEs have this "to gather and organize the internet's information" as a business statement in one form or fashion. Also, their technology for doing so has gone way past what it was 10 years ago and is constantly improving. C'mon! Do the work! Yes it's a lot of work, but anything of value always is.

#19 Jill

Jill

    Recovering SEO

  • Admin
  • 33,244 posts

Posted 03 November 2007 - 01:16 PM

Awesome. What you're saying confirms exactly what I've always thought. To me, it's very much common sense.

Of course, it would probably be worth testing something similar, but with only 1 or 2 words in the URL instead of your much more "spammy" method! smile.gif

#20 qwerty

qwerty

    HR 10

  • Moderator
  • 8,695 posts
  • Location:Somerville, MA

Posted 03 November 2007 - 01:55 PM

But you're treating keyword stuffing and keyword inclusion as if they're the same thing. So you have data comparing domain.com/dksldjr09.htm with domain.com/the+canon+powershot+s110+digital+elph+camera.htm, but what about domain.com/electronics/cameras/digital/canon/s110-elph.htm?

#21 BBCoach

BBCoach

    HR 5

  • Moderator
  • 402 posts

Posted 03 November 2007 - 02:48 PM

Jill,

QUOTE
Of course, it would probably be worth testing something similar, but with only 1 or 2 words in the URL instead of your much more "spammy" method!


It's not my spammy method. It's the marketing geniuses holy grail spammy method. I don't even want to go there because I'm convinced that taxonomy of a site is a major factor and not the URL.

qwerty,

QUOTE
domain.com/electronics/cameras/digital/canon/s110-elph.htm


I quit building sites like that back in 2001. The log data was not nearly as impressive as it is having everything showing up at the root level. With that scheme it took weeks or months for products to get into the top 20 rankings in the SEs. Now I've got it down to roughly 3-10 days. Been there. Done that.

My primary goal has always been to get products ranked in the SEs in the shortest amount of time and keeping them there. I used to be a black hat SEO and have tested and tried every conceivable iteration to accomplish my goal. However, once I realized what the goal of SEs are and branded that into my soul, and saw first hand that tricking them wasn't going to work because they continually improve their algo, I was reborn into a white hat SEO. Besides. It's logical and easier than constantly trying new things. And it works fantastically.

QUOTE
But you're treating keyword stuffing and keyword inclusion as if they're the same thing.


Ok. Including keywords in the URL from my analysis is a waste of time and a person should focus most of their efforts in properly categorizing their web site. If you don't think so, then please continue doing it. I'm convinced it's a waste of time and in two months some marketing folks will be shown that it is. Question is how am I going to fix that one category that has more than 3000 products in it? Ain't figured that one out yet, but will prolly just bite the bullet and those products will be put back another 6-9 months in their rankings.

#22 Jill

Jill

    Recovering SEO

  • Admin
  • 33,244 posts

Posted 03 November 2007 - 09:21 PM

QUOTE
It's not my spammy method. It's the marketing geniuses holy grail spammy method.


Sorry, didn't mean to suggest it was your method, was just meaning the collective "you." smile.gif

#23 BBCoach

BBCoach

    HR 5

  • Moderator
  • 402 posts

Posted 03 November 2007 - 10:45 PM

Jill,

I figured that's what you were saying and it's not a problem, but I ain't taking any blame for that stupid idea. I'm placing it squarely on the marketing team's shoulders. Hopefully they'll either be fired or at the least not listened to concerning SEO. 250 man hours were spent making their crap work. Marketing people should stick to user interface issues and issues concerning the completion of a sale. These guys have come up with some of the stupidest SEO crap you can imagine. Most of which I've already tried and have proven that it fails. I'm the pooh pooh man in their opinion. That blog by Matt has caused more confusion than it's worth. I reckon it might have relevance where there is no taxonomy on a web site, but that's a juvenile site design at best. I have NEVER dealt with hearsay other than to test its validity in a real-world scenario and either implement it or delete it. My job is to get results. Period. I don't get the luxury of trying new ideas without results. It's real hard to tell the owner of a company. "Sorry, but I tried this and just had your web site's index DELETED from Google." I'm through with TESTING new ideas. It's either my way or the highway. I'm beyond caring any longer and will take my skills else where if necessary. And, yes qwerty this is personal for me, and I can PROVE that keyword inclusion in a URL is BS.

#24 qwerty

qwerty

    HR 10

  • Moderator
  • 8,695 posts
  • Location:Somerville, MA

Posted 04 November 2007 - 10:14 AM

Well, it's not personal for me, so I won't argue. For me, it's common sense, usability, and a believable explanation from a usually believable source.

#25 webd

webd

    HR 2

  • Active Members
  • PipPip
  • 46 posts

Posted 24 December 2007 - 08:01 PM

QUOTE(BBCoach @ Nov 3 2007, 01:12 AM) View Post
I can PROVE it doesn't work. Period. No referrals = No sales. All referrals are to product pages NOT to the "test links." Period. Can you prove otherwise with your sites?


Google, and possibly other search engines, read keywords in URLs. That doesn't mean that you will rank for those keywords. You can also get very high rankings even without an h1 element or properly optimized title. You can get higher rankings by optimizing those aspects of the site though.

Here is an example showing that Google reads keywords in URLs:
http://www.google.co...h?q=content1857

I wouldn't change existing URLs to add keywords, but if starting from scratch, or rewriting URLs for another reason, I would incorporate keywords.


#26 chrishirst

chrishirst

    A not so moderate moderator.

  • Moderator
  • 7,718 posts
  • Location:Blackpool UK

Posted 25 December 2007 - 06:20 AM

For a search that has only 28 results and a line on the cached page that says

QUOTE
These terms only appear in links pointing to this page: content1857


You are mixing cause and effect up

there will be a link somewhere to those pages using the URL as the anchor text, possibly a scraper site or more likely a republished RSS feed

#27 webd

webd

    HR 2

  • Active Members
  • PipPip
  • 46 posts

Posted 25 December 2007 - 07:01 AM

QUOTE(chrishirst @ Dec 25 2007, 07:20 AM) View Post
You are mixing cause and effect up


I don't think so.

QUOTE(chrishirst @ Dec 25 2007, 07:20 AM) View Post
there will be a link somewhere to those pages using the URL as the anchor text, possibly a scraper site or more likely a republished RSS feed


How are you eliminating the possibility that Google is reading keywords in the URLs? Google has even highlighted the keywords in the URLs in bold. Matt Cutts has specifically said that keywords in URLs are one factor that Google uses in its algorithm. Show me some proof to refute the very clear indications that Google reads keywords in URLs. I was just using that one search as an example. I regularly see this happening for non-competitive searches.

You can also test it by registering a new domain. Make a page with a unique name like example.com/1329djuioer293.html. Then get an inbound link to the home page of the site. As soon as it is indexed, search Google for your unique phrase and Google will return a match based on the keyword in the URL.

BTW, RSS feeds don't create HTML links. RSS uses a <link> element (not an <a> element), and uses HTML entities to escape the content of posts. I've gone through all copies of that page on the Web and there are no IBLs with that link text:
http://www.google.co.....e an advance"



#28 chrishirst

chrishirst

    A not so moderate moderator.

  • Moderator
  • 7,718 posts
  • Location:Blackpool UK

Posted 25 December 2007 - 07:34 AM

QUOTE
, RSS feeds don't create HTML links. RSS uses a <link> element (not an <a> element),


Correct!

Which is why my comment says
QUOTE
more likely a republished RSS feed



QUOTE
I've gone through all copies of that page on the Web and there are no IBLs with that link text:

Sure, but have you been through every scraper, feed republisher, article scraper/grabber etc etc

And in fact, from one of your example searches

HTML
<div align=center> <a class='dkl' href='http://www.grinbold-jodag.de/content/1857.html'><strong><b style="color:black;background-color:#ffff66">content1857</b></strong></a><br></div>



On the bolding, it has been established long ago that the bolding of the words in the SERPs is simply a display feature and not to be confused with ranking factors.

#29 webd

webd

    HR 2

  • Active Members
  • PipPip
  • 46 posts

Posted 25 December 2007 - 07:41 AM

QUOTE(chrishirst @ Dec 25 2007, 08:34 AM) View Post
And in fact, from one of your example searches


Which page is that from?

QUOTE(chrishirst @ Dec 25 2007, 08:34 AM) View Post
On the bolding, it has been established long ago that the bolding of the words in the SERPs is simply a display feature and not to be confused with ranking factors.


How was this established? I would venture to guess that most SEOs agree that keywords in URLs are a factor. A Google engineer has even confirmed this. If you try my experiment with a few domains you will see. Get them through 1and1.com and the experiement will cost you less than $20.

Like I said, how are you eliminating the possibility that Google is reading keywords in the URLs?

#30 Jill

Jill

    Recovering SEO

  • Admin
  • 33,244 posts

Posted 25 December 2007 - 01:05 PM

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? Is it similar to the weighting Yahoo puts on the information contained in the Meta keyword tag (barely any, but if it's the only place a word appears, and there is little-to-no competition, it will be weighted). Or is it given a lot of weight, like the Title tag (one of the most highly weighted areas on the page).

If it were the latter, certainly it would be worth changing URLS to use keywords. Personally, I don't know what the exact weighting is, but from my observations (not tests) it seems closer to how Yahoo weights keyword meta tag info. But this is just my opinion, not any sort of fact.




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.