structured URL's or non structured? Whats best?

[doodledunk]

Hi all, need your worldly advice.

I'm working on a very large ecommerce site with lots of duplicate product pages. The developer wants to create a canonical tag which removes all the hierarchical information e.g.

www.example/category/subcategory/product1a.html

becomes

www.example/product1a.html

and then he's going to change internal linking so that all product pages no longer carry the hierarchy info.

Is this a good idea for seo or not? What is best practice?

Thanks Doodle

[goodroi]

I would prefer to use one version of the pre existing duplicate page instead of creating a new version of the page but it would also depend on scale. What scale are we talking about 10 pages or 10,000 pages?

[levo]

I would go for www.example/product/1a.html

[lucy24]

and then he's going to change internal linking so that all product pages no longer carry the hierarchy info.

I think that's the most encouraging line. Link to the canonical, not to something that will be redirected.

A form like

example.com/pagename.html

seems a little risky for a large site. Can you be absolutely certain you'll never have two products in different categories with the same name?

[Robert Charlton]

and then he's going to change internal linking so that all product pages no longer carry the hierarchy info.

That's the best phrasing I've heard to describe the technique. Yes, that's an excellent idea.

Can you be absolutely certain you'll never have two products in different categories with the same name?

To be safe, you probably should have an ID number that's unique to each page in your urls... but as part of the page filename, not in a separate ID in an ID number directory.

Look carefully at what g1smd says in this thread...

Choosing the best url structure
http://www.webmasterworld.com/google/4475535.htm [webmasterworld.com]

...and also in this thread, which I link to in the above, in which he posts examples....

How important is it to organize pages into directories?
http://www.webmasterworld.com/google/4364322.htm [webmasterworld.com]

While the old way was to have a folder heirarchy for the site and show that structure in the URL, nowadays that will often get you into trouble....

[doodledunk]

Thanks for all your help! The site is on the large side (about 8000 pages) of which about 5000 are unique. There are lots of duplicate product pages in sections like 'clearance' and 'special offers' so the idea is to eradicate the duplication by removing breadcrumbs from the URL and just putting files all at root (or there abouts) level.

Does this video : [youtube.com...] still stand up in today's algorithm? Should I try and put in descriptive keywords into the filename? (rather than file structure?)

I have looked through lots of material from matt cutts and cant find a conclusive answer to weather it's better to create a vertical / silo / directory structure for a website OR weather it's better for web pages to exist closer to the root directory? OR if in fact it matters either way.

Does anyone else know?