Discussion about SEO Friendly URLS

[calguy1000]

Here's a question......

Which is better for Search Engine Optimization:
URLS like this:  www.mysite.com/News/article-name

or this:  www.mysite.com/News/1234
(where 1234 is an integer id that uniquely identifies an article).

My contention is that it doesn't make a difference....
I understand that urls like this:
http://svr.techcom.dyndns.org/cms_test/ ... eturnid=39
are totally unfriendly and hard for the spiders to crawl through.

But for the top two URLS... I can't see a difference, and my research tells me that there shouldn't be one.

Can somebody please tell me why the first url is better (and not from a human point of view) for SEO optimization?

[cubix]

I believe the top one (from a SEO point of view) would be the prefered as it offers keywords in the pagename.

If you would like a method of uniquely identifying the article, why not combine the two or use the date and name format..

eg 2008/04/01/article-name

[calguy1000]

All I can tell from my research is that, other than human factors, there is no proof that
www.mysite.com/News/my-puppy-died is any better than www.mysite.com/News/1234 as far as search engine ranking goes.  I've spent the last 1/2 hour with google trying to figure this out, and it all appears to be conjecture.

There may have been a time when keywords in the url were important, but I don't think it matters now.

As far as Date/name format..... as far as search engine ranking goes I can't find anything at all on that.  And a url such as:    2008/04/01/article-name implies to me that 2008/04/01 should show all of the articles from that date, and 2008/04 should show all of the articles from that month, and that 2008/ should show all of the articles from that year.  My experience tells me that that doesn't work on many sites (only a few as far as I can see, and I haven't put a pattern on it yet)... and that most of the sites return 404 errors when you try to do this.  Thereby, those URLS aren't even people friendly.

[Nullig]

There are lots of SEO "myths" and outdated information out there, which a lot of unscrupulous companies use to pretend to help you promote your site and I think this is one of them.

Way back when search engines began, it may have been true, but the advanced spidering techniques and algorithms that they employ today put a lot less emphasis on the "friendliness" of the link and more on the content and traffic. Add to that the fact that companies can "buy" rankings from these search engines, as well.

Strategies like "stuffing" keywords and creating false "hidden" linkages became so widespread a few years ago, that the weight of these things in the overall ranking consideration dropped dramatically. Yet, they are still touted by some, and believed by most, to be the way to "fool" the search engines into giving you a higher ranking.

Google, along with the other major search engines, has become much more sophisticated in their assessment of site content.

Nullig

[Dr.CSS]

SEO likes a clean URL, page ranking on search engines has to do with links to your site, more pages that say 'hey look at this', and content, searching has to do with Content...

If you have metadata tags with 100+ keywords to try and get in the top of the search pile you will just shoot the foot, and actually get punished for it...

Spiders crawl and read the content, opening and following links, I found my site in a search and the link was to an open news item from the side bar, hence the problem with tables (the redundancy of table, td, tr, drives them nuts) I've converted table based layouts for $ because that was the first thing their high dollar SEO's told them to do...

Anywho, Reapeated phrases/words in content, that mean something, will got you to the top faster than mega keywords, unless you pay for it, did research for a fellow who had t-shirt site w/ 350 keywords in and after sifting thru 100+ pages on google I gave up trying to find his site, and he paid $$$$ to have his SEO optimized...

A URL with words will help to the degree that when it's searched/googled it will be part of the findings that the human will see, search engines don't care they are just looking for words/strings of words to index...

[giggler]

I'd have to say that www.mysite.com/News/article-name or www.mysite.com/2008/1/article-name makes a difference. This is from having a site with subpages as /234/23/page23.html changed to category-name/article-whatever-name.html and it started to show up on the first or second page of google based on searches for something like "article whatever name" -- so I definitely think that it makes a nice difference with the addition that it is nicer to look at for human as well. This of course also has to do with other variables such as page title (important), heading title, content keywords (and not over using the same keyphrases), backlinks, keyword anchor links as well as overall site keywords/phrases.

Since search engines looks at a site as a whole, if you have /sub-keyword/article-names-that-may-include keywords, when you link to pages like that, obviously the keywords are there as part of the link, which helps also individually as it's own page AND search engine will see more keywords overall in the site as a whole...make sense?

SO anyway...it's a combination of everything, but since I've always used article-name and have done well especially with regional SEO - I strongly believe it is very beneficial to have this.  Searching for keywords yielding results with 3423523.html of course can happen, this is especially true for big sites that already has good backlinks (keyword may also be before the ###.html). For smaller sites that needs more help..it will help:

Just for fun search for "shopping in alaska" and see the results that's with 34232.html but also with keywords/in/the/url.html

and try "using seo friendly url" and see the results... and it's an obvious choice to use article-name rather than random####....

did research for a fellow who had t-shirt site w/ 350 keywords in and after sifting thru 100+ pages on google I gave up trying to find his site, and he paid $$$$ to have his SEO optimized

I just read this and had to comment on it =). Having that many keywords in head tag is overstuffing - may have worked in the old days, but obviously not anymore. It is recommended to use only about 5-10 target keywords/phrases per page for keyword tag. Not much, but that's why it's targeted...

2008/04/01/article-name implies to me that 2008/04/01 should show all of the articles from that date, and 2008/04 should show all of the articles from that month, and that 2008/ should show all of the articles from that year

The use of the year/month/date to me never meant the above. I would not try and go to 2008/ or 2008/04 on a site that had it. I think of it as more of a way to differentiate the article or giving it unique id in some ways along with it's article names. This is because using only-article-name can get confused when you have two of the same only-article-name, so the year month date is easy way to separate that. articleid#/article-name will also do that (in a mode_rewrite kind of way), but in wordpress/blog world, dates are more used because it also identifies when the article was published and I don't think it was a form of organization hierarchy.

Sorry this is so long - also why it would be important in using it for news modules and blogs?! It's because news and blogs are content that are updates "more frequently" and search engines like fresh content. Pretty url in content is not only for human eyes and if you would do this for regular content, it's only natural to want to do it for news/blogs...

[calguy1000]

Thanks.... your answers make alot of sense and mostly re-confirm my oppinion.... my current supposition is that keywords in the url don't make 'much' of a difference, but they can make a difference in a well organized, well thoughtout (for SEO purposes) site.

I think of it as more of a way to differentiate the article or giving it unique id in some ways along with it's article names. This is because using only-article-name can get confused when you have two of the same only-article-name, so the year month date is easy way to separate that. articleid#/article-name will also do that

This subject came up because of the Calendar, News, and CompanyDirectory modules that I've been working on of late.  I noticed that year/month/day/article-name didn't work (so I fixed it) in the Calendar module.... and then noticed that year/month/day were just ignored anyways, which didn't make much sense to me.  I also had the thought that nothing in the calendar, companydirectory or the news module prevent you from entering the same article name more than once.  Therefore, the article-name was inssufficient.

Maybe something like calendar/ID-article-name would be useful as the article-name portion could be ignored for the purposes of navigation, but the article name (or an alias generated from the article name) could still be used to assist in SEO optimization.  However, I view this as a 'minor goal'.  Shorter URLS without all the variables on the GET line would be the primary goal because Search engines definately assign lower scores to long, complex URLS.

[giggler]

Maybe something like calendar/ID-article-name would be useful as the article-name portion could be ignored for the purposes of navigation, but the article name (or an alias generated from the article name) could still be used to assist in SEO optimization.

I also had tried category/article-name-id before and ended up using something like category/id/article-name because it looks better and doesn't have the extra id number in the way for seo.

So calendar/2324/article-name.html will "look better / look more organized" than calendar/2324-article-name.html to human and to spiders.

[Pierre M.]

Hello,

I'd like to share some blahblah on this topic :

I prefer to say User friendly URLs rather than bot friendly URLs. Because I think if a URL is human friendly it is easy to wget by any bot.

I prefer /boat.html and /car.html because I think they are more human usable than /2615.html and /321.html.

I like URLs like http://w3.somesite.org/blog/2008/01/02/ ... e-123.html and shortcuts like http://w3.somesite.org/blog/2008/01/02/ or http://w3.somesite.org/blog/2008/
They are doing pretty well among http://w3.somesite.org/events/2008/03/1 ... words.html and http://w3.somesite.org/aboutus/customers.html

I agree the shorter URLs are better. Things such as .org/cms_test/svn/index.php?mact=News,cntnt01,detail,0&cntnt01articleid=6&cntnt01origid=15&cntnt01returnid=39 are ugly.

So my opinion is : let us make nice websites with human friendly short URLs, it isn't more expensive than making ugly ones, this will please our target audiences and may be the bots too.

Pierre M.

[calguy1000]

Sorry pierre, but this topic is about URLS that are SEO friendly.... lets keep the human factor out of 'this' discussion.

[Pierre M.]

Sorry, my mistake. I've misunderstood.

Pierre

[Courty]

Hi Guys,

Intresting topic...

I believe human friendly URL's do make a difference to Google and the other SE's as mentioned above.

Try typing in BMW to Google, not only do you get http://www.bmw.co.uk but you also get sites with http://www.mysite.com/[b]bmw[/b].html in the page part. (in Google it's even highlighted in bold for you !)

This is also true of the Adwords paid for ads (look in the right of the Google results page), having a key word in your url does get picked up.

I agree it doesn't appear to matter if it's after a \news\general\bmw-story.html type link other than if your using adwords with a 25 char limit (then I use Alias's anyway)

Bottom line is that I believe having key words in any part of the path helps, the most points are for having it in the actual URL followed by the page name, title, content etc..

Courty 

[myshko]

My 2 penneths worth...

I agree with most things said here, SEO is one of those messy grey areas, and sometimes there isn't always a real black or white answer.

When optimising for SEO, I draw from the content always, in that whatever the page is, and whatever it contains, I try to communicate this as simply and effectively as possible to both bots / spiders AND human users alike, because whilst a bot may be ranking my page, a human user does the same thing when presented with a list of potential pages, you scan urls and titles and look for ones that makes sense or seem relevant.

If I search for the solution to a problem, I quickly filter whole pages for keywords I'm interested in.

Any way you can communicate what a page contains and it's use to me is good SEO, that way you plan for the future, just because a search engine ignores it now, does not mean it will in the future and vise versa. Google changes up its algorithms often.

This means optimising your domain name, urls, page titles, meta keywords, meta description, page links, image alt tags and page content for each important page. It's not about tricking the engines, just make good clean content and a nice organised site and the rest follows in time.

So, if you can have urls which say 'http://some-site.com/news/article-title/' all the better, as long as:

URLs are short, simple, are relevant to the page content and preferably terminate at the directory level '/'.

Good topic

D