Database URLs

[CaseyC]

I have a database driven site that I ahve been working at optimizing for several months now and have had varying degrees of success. Most of my success has come with more product specific keywords and less with the general terms. As I have been studying our site and the way the urls are created through the database, I am wondering if the way my URLs are created could be an issue. For instance, my product pages may go something like this:

domain.com/productinfo.php?sku=100-1&url=oldbluecars

While the section pages, where groups of like products are listed look like this:

domain.com/sections.php?section=(section name)&surl=(url name)

What I see has happened is that if I have a Section named Blue Cars and the URL as Blue Cars, the URL on the site shows as:

domain.com/sections.php?section=BlueCars&surl=Blue Cars

I am thinking that this may actually look spammy to the search engines as I inadvertently have the same keyword in the URL twice. Could this be why my product pages seems to do OK, but many of my Section pages that end up with URLs like this do not? Would I be beter off not having the url piece at the end or naming it something different than the Section name?

[AussieWebmaster]

I would program individual page name as the url name except where the page is index.html of a section... use section title

[beu]

In addition to aussie's advice, I'd set up a Google webmaster account to see what links to the urls in question and how many there are.

[CaseyC]

As I have been looking at this further I am seeing that it appears that my section pages are being indexed with different names that what they should be, in short it appears that every section page has 2 urls. For instance one page should be named www.example.com/section=cars&url=car-parts, instead what is being indexed is www.example.com/section=cars.

Both URLS go to the same page, however one has a page rank of 4 and the other 0. I see this has happened to most all of my Section pages. I assume this could be problem and could be hindering the rankings for these specific pages. COrrect?

[AussieWebmaster]

Casey one page is seen as the canonical url the other seen as a copy

[sitetruth]

As I have been looking at this further I am seeing that it appears that my section pages are being indexed with different names that what they should be, in short it appears that every section page has 2 urls. For instance one page should be named www.example.com/section=cars&url=car-parts, instead what is being indexed is www.example.com/section=cars.



The standard format for URIs with parameters is

http://www.example.com/path?key=value&key=value&key=value

This isn't strictly enforced by browsers, and there are major sites with totally different syntax, but because your URL doesn't match any standard formats, Google isn't parsing it your way.

[CaseyC]

Aussiewebmaster...how do I correct this? Do I need to set up redirects so that only 1 page is truly indexed? In many cases the URL that has been indexed is not the URL with the higher page rank.

[beu]

If using a 301 redirect is an option, I'd consider it for sure. From the sound of it there is no malicious duplicate content here but chances are you're still diluting relevancy and PR.

[CaseyC]

The PR is my concern. The URL that Google has indexed has a PR of 0, while the page that is intended to be indexed has a PR of 4. When I check that stats on the PR4 page, it shows it as not even being indexed and has no cache. So if I can do a 301 redirect, do I redirect the indexed page to the higher PR page or vice versa. This is my first experience with this issue, so I am still getting my mind around it, but it really does make sense as to why my product pages rank well, but my section pages do not.

thanks

[cryptblade]

CaseyC, you definitely have a dupe problem - multiple URLs going to same page, Google indexing the wrong URL and incidentally, the one with no authority.

If you can, you should do a URL rewrite. Depending on what your site is built on (php, asp, .net, etc) there are specific resources out there to help you do that.

You'll need to make sure your pages all reconcile to the correct, authoritative URL, and you'll need to map out your internal linking architecture. It sounds like your site actually links to the same pages in multiple ways (is why different ones are indexed) so you'll need to fix that so that each page is linked in only 1 way (1 standard, canonical URL)...for each page.

...I don't envy the work you'll have to do.

[Jazajay]

I tell you SEM's.
Heres the programmers creative solution rather than hacking through with a re-direct. ;)

www.example.com/section=cars&url=car-parts (http://www.example.com/section=cars&url=car-parts), instead what is being indexed is www.example.com/section=cars (http://www.example.com/section=cars).
Can you explain why they both go to the same page?

Why is the car-parts varaible needed?

The way I look at that url is that in fact car parts is a sub category of cars. Unless car parts is the only sub category, and you don't plan on adding any more in which case car parts should be the category.

This is how I would do it. I'm missing key information what your site caters for being the main one.

note: example.com is your site name.

example.com/category.php?section="cars"

This would just be the category with links to any sub categories car parts being 1.

<title>Cars Category</title>
</head>
<body>
<h1>Car Category</h1>
<p>
Please chose a category to go to -<br />
<a href="car-parts.php">Car parts</a><br />
<a href="car-interior.php">Car Interiors</a><br />
<a href="car-seats.php">Car Seats</a><br />
</p>
The have the sub category car parts under the car category.
The url would then look like this to a separate page -

example.com/sub-category.php?category=cars&subCategory=car-parts

then on that page you list the products in the car parts database.

If all you sell is cars and car assecories then it would just be

example.com/category.php?category=car-parts


This would be less work, ok maybe not. But this to me is a lot better than any redirection.

Solution 2 -

Write logic that places the meta noindex on the page you don't want indexed. This case I would say -
www.example.com/section=cars (http://www.example.com/section=cars&url=car-parts),
Should get the meta.

So something like -

<?php
if('http://www.yoursite.co.uk'.$_SERVER['REQUEST_URI'] != 'http://www.yoursite.co.uk/section=cars&url=car-parts)
{?>
<meta name="robots" content="noindex" />
<?php }?>
This tests to see if the url is the one you want if it isn't it gets the meta noindex but all links from the page will still be followed.

If the section is dynamically populated this will need to be modified let me know. 2 second Job.

Let me know if you need help with any onpage re-direction if you decide a hack is the way to go.

Jaza

[CaseyC]

well, what was programmed was the string that followed the section name was added as a way to include a few more keywords. However what I was afraid was happening was that having the section called say "Car Parts" and then having the url string at the end say something like "car parts" again was causing my urls to look spammy. Not all were looking just like that but the similarities between the actual Section name and url string was close enough to still cause me worry. What I have discussed with my programmer, based on feedback here, is to just do redirects from what appear to Google to be copy pages to the the pages they see as canonical. In doing this we will be doing away with that extra url string and the url will simply end with the category name i.e. www.domain.com/section=carparts and just keep the urls short and sweet. As I said, fortunately this has only happened on a handful of pages and is not going to be a tremendous undertaking, so she says.

This site is my first database driven site to maintain, so I am going to chalk this error up to a learning experience.

[Jazajay]

Ok fair enough, at least I get to show off my creative problem skills again :).

Redirect, sorry hack :D, it to car-parts not carparts though.

Jaza

[AussieWebmaster]

well, what was programmed was the string that followed the section name was added as a way to include a few more keywords. However what I was afraid was happening was that having the section called say "Car Parts" and then having the url string at the end say something like "car parts" again was causing my urls to look spammy. Not all were looking just like that but the similarities between the actual Section name and url string was close enough to still cause me worry. What I have discussed with my programmer, based on feedback here, is to just do redirects from what appear to Google to be copy pages to the the pages they see as canonical. In doing this we will be doing away with that extra url string and the url will simply end with the category name i.e. www.domain.com/section=carparts (http://www.domain.com/section=carparts) and just keep the urls short and sweet. As I said, fortunately this has only happened on a handful of pages and is not going to be a tremendous undertaking, so she says.

This site is my first database driven site to maintain, so I am going to chalk this error up to a learning experience.

You seem to be getting the idea... if you need specific answers post away... this one is helping newer members pick up some insight...

[Jazajay]

Yeah I totally agree Aussie.
Dynamic sites are totally different and do require a special treatment.

....so I am going to chalk this error up to a learning experience.
I do it every day TBH.

But there's never a stupid question, well unless I'm asking it that is :D

Jaza

[CaseyC]

Well, we got the re-directs done, but I really have not seen much improvement in rankings for a few keywords I have been looking at. Most of the pages have been crawled and appear to be indexed correctly now. ON thing I see is that I have lost some page rank. With a change like this, could I be looking at several weeks beofre I can really gauge the success. Will it take Google a while to get all of the re-directs clear in their system and then update my Page Rank?