Tried searching here but no results. So, does it matter or no? Example <a href="http://www.cursors-4u.com/">Home</a> (Absolute) <a href="/">Home</a> (Relative) I read a few blogs, but they were outdated. Some said it mattered, some said it didn't matter. Some said, it's only to save, like 00005 kb of bandwidth or some crap like that. What's your thoughts and experience on it?
Fact #1: In nearly all cases it's easier to develop on a test server using relative URLs since it's linking /within the directory you're working from. Fact #2: Feeds and code pulled from a page often does not correctly pull the internal links within the content causing dead links in the case of relative URL linking. Fact #3: Absolute URLs require more text adding to the file size of your pages. Theory #1: Absolute URLs more often translate to to actual domain credited links in the eyes of SEs. Theory #2: If your domain has important keywords in it, absolute linking makes your content more keyword rich. So with that said, make your decision . My opinion is the only real negative to absolute is the larger file size, but it's so little of a difference I don't see it causing any problems, however developing can be a pain. So, I typically go with relative during development and find&replace to absolute when going live.
Theory #1 not sure about, but Theory #2 makes a whole lot a sense. I guess with that a lone...should be more than enough for me to go absolute url now. The only problem is the www, and non www
If you plan on doing any PPC content advertising I recommend www. When you put your domain in the ad, you of course put your preferred version as the display URL. If you put in www.domain.com and it's picked up as a content ad on a website like a news source or another site that likes to take a daily or weekly snapshot of their website to pull into a .pdf, with the www it will import as a link and you get a linkback, without the www it will just be text. This happens more often than people realize. If you know a site that does decent volume PPC advertising with content enabled, do an advanced google search for .pdf with their "domain" in the search. One of my sites has over 80 results of .pdf files that happen to have had our adwords ad on it at the time of creation. It's filed even by Google as a linkback now . Yet Google says PPC doesn't affect SEO
Oh, and to further explain Theory #1, when the SE spiders the site and sees <a href="www.domain.com/blah.html"> on site link credit is given to "www.domain.com/blah.html" If it's <a href="/blah.html"> the theory says in some cases it's filed as just "blah.html" by some SEs and on site link credit isn't given fully to the page you're linking to. Just a theory one might prove or disprove by combing through all SEs indexed pages and backlinks on file... or might not since not all will give you full reports lol.
The search engines treat absolute and relative URLs identically. They have to reconstruct the complete URL from relative URLs in order to use them. The only disadvantages in using relative links are (1) if you make a mistake, you create a bad link that doesn't help your rankings and (obviously) inconveniences users, and (2) if you have a canonicalization problem (ie. your site is being indexed both with and without the "www." subdomain prefix), the use of relative links can make the problem worse. There's no "style points" for using absolute links. The value of having your keywords in your domain name only comes into play in terms of links is when the anchor text is the URL of a page on your site, not in the content of the href attribute.
The SE can not reconstruct your URL if it's crawling your content from another site that scraped your code or pulled a feed you have out there (as one example where absolute gets a link giving you credit for your content that relative doesn't). This is regardless of if they took it with or without permission. You write as though this has all been proven hands down, but it hasn't not even in the slightest. The theory that some SEs will credit the on page content and wording as being more keyword rich with absolute URLs when your domain has an important keyword is a very understandable theory. Whether I think it's the case or not with the big SEs, I don't feel as though any of us are qualified to discount it in all cases and SEs without any documented testing proving otherwise from all SEs. This is along the same lines of some out there using line out code to get their keyword/phrase on page in the code a few more times. I consider it hidden text and blackhat, but many are doing it feeling it helps.