Absolute vs relative URLs in SEO

[debbidoo]

Hey there, SEO bods

Just wondered whether it makes any difference to SEO to use absolute URLs in internal links, instead of relative URLs?

e.g.

is http://www.website.com/website-website.html better, worse or no different to ../website-website.html? I'm talking about links within content (e.g. a big chunk of text where you mention one of your products or services and link to it mid-paragraph) - and also, in navigation links (which appear on every page).

Looking forward to seeing all your different opinions - I think the SEO forum is my favourite part of UKBF because there are always so many differences of opinion, all of them interesting

Debs

 

[awebapart.com]

You'll certainly get differences of opinion on this one.

Where possible, I usually go for relative urls. Mainly because it cuts down on page file size and loading time, it is more flexible and efficient, and it reduces code to content ratio.

 

[debbidoo]

You'll certainly get differences of opinion on this one.

Yes, I thought I might

I usually go for relative urls. Mainly because it cuts down on page file size and loading time, it is more flexible and efficient, and it reduces code to content ratio.

Excellent points - thank you

 

[UKSBD]

absolute can benefit if your pages get scraped

 

[debbidoo]

absolute can benefit if your pages get scraped

Very true - and one of the reasons I've usually used absolute URLs in the past, as I've read a few stories on UKBF of people having their entire websites duplicated by competitors or whatever. Amusing when the thieves forget to change the links though - I remember that happening to someone here last year

 

[UKSBD]

Better when they leave your statcounter code on the pages, even better
when they have the site copied on their desktop in a folder and have their
PC or a folder named after them or their company

 

[awebapart.com]

Scraping/content theft is one of the arguments for absolute urls, although personally I think that any good web scraping software would be removing absolute urls by now. So the threat nowadays, where absolute urls offer an advantage, is probably more from the not so careful manual user copy and pasters. In some cases the chance of a back link might be good, in other cases you have to ask yourself do you really want a backlink from a site that just copies other sites? Is there a chance of being associated with a bad neighbourhood? You also have to ask yourself what is the inbound link value of a duplicate content page? Will the action of pasting into a CMS/blog/forum remove the HTML links (or nofollow them) anyway? It also depends on what you do when you notice people copying your content, do you embrace it, or do you try to stop copyright infringement?

If you are submitting articles or providing RSS feeds to content, then again, absolute urls might be more appropriate.

With this whole debate I should add that in the grand scale of all things SEO, I personally do not think the absolute vs relative debate is that important for SEO, it is definitely not a major ranking factor.

One thing I should say is that there are different types of relative urls, page relative (relative to the current page), and site relative (relative to the base directory of the site). I usually go for the site relative urls, starting with a / and using the BASE HREF HTML tag to define the base url. Page relative urls (starting with a name or ..) have their uses too, but they can cause confusion if you move pages around, and can sometimes cause spiders to get stuck in loops (indexing contact then contact/contact then contact/contact/contact etc).

Here is something I sometimes do when there is a debatable SEO issue. Have a look at the top 10 sites for the search term SEO, to see what they do. Do they use relative or absolute urls for internal links:

1. en.wikipedia.org... relative
2. justsearching.co.uk absolute
3. seo-london.com relative
4. ihaveawebsite-nowwhat.co.uk absolute
5. google.com... absolute and relative
6. seo.com absolute and relative
7. freshegg.com relative
8. seochat.com absolute and relative
9. seoco.co.uk absolute
10. searchengineoptimising.com absolute and relative

So there is a mix of opinions in that top 10. Some of those sites with a mix of both actually have absolute urls for navigation area links, and relative urls for the main content (which kind of ignores any content scraping benefits).

Out of the many arguments for and against relative and absolute urls, the biggest myth I've heard so far, from Bing's webmaster team nonetheless, is that relative urls do not pass on link juice. This myth is nonsense and easily disproved, at least for Google link juice, by checking things like the page rank of a section link of a high PR site, a page that is only linked to internally via relative urls and not linked to externally. (BTW there are also sites in Bing's top 10 results list for SEO which use relative urls).

 

[debbidoo]

What a brilliant answer, Paul - thanks so much

I think that given everything you've said, if relative URLs don't appear to carry much less SEO weight than absolutes, then it seems the benefits of trimming down the code would make relative URLs the better choice.

 

[FireFleur]

Most HTML documents are sent with compression, so there is not that much advantage in keeping the URLs short, it is often quite negligible in term of overall code size in transmission.

One of the advantages of using absolute URLs is if you mix domains or protocols and have repeated elements such as the navigation. If you move to an HTTPS server, say for contact or basket, then even if that is an absolute URL if the remaining navigation is relative then you stay on the HTTPS rather than go back to the HTTP, that creates a duplicate content issue with the site, and effectively reduces the amount of link juice to some pages i.e. the ones linked off the HTTPS pages.

 

[awebapart.com]

I think that given everything you've said...

I wouldn't go just by what I say, after all it is just my opinion, and I could be missing something SEO wise.

Like you, I was hoping that more of the UKBF SEOs would chip in with their views and reasoning on this subject, since unlike other SEO discussions, this question of SEO benefits of absolute vs relative urls is pretty rare on this forum.

 

[FireFleur]

Relative is easier to develop in, as you can have many servers without having to alter the URLs for each one, the entry point is domain.

A development, staging and live server could all have the same code with relative, but if you go absolute then you have to change the code for each one, which generally means multiple templating.

You will sometimes see a hybrid approach (repeated elements are absolute, content is relative), but even with the extra overhead of multiple templating absolute is easier to work with to test a site.

 

[Lmb Supplies]

use absolute better for seo