May 2, 2008 @ 7:25 pm in Uncategorized by   8

Linking to Related Websites Helps Google Ranking – A Definite Proof

It is a well known fact that receiving links from related websites or web pages can help with rankings, but what most SEO gurus would not admit ( knowingly or unknowingly) is that linking out to a related web page or website can also help your website ranking.

Recently I have set up an experiment page to test some of my assumptions about Image optimization but before I could come to any conclusion on that I got this definite proof of how the pages to which you link from your website can affect your rankings.

What I Did

  • I created a HTML page that contains two images with the same Alt attribute on both of them. I intentionally did not put any Title or Meta tags or any text on that page. Then hosted the page on one of my client’s server ( with his permission) which hosts his website which is no way related to the images.This page with the images was an orphan page without any link to the site.
  • I ensured that there is no search result in Google for the nonsense words that I used as Alt attributes for the images
  • I set up a link to this image page from a blog with a simple text that is no way related to the text used as alt attribute.

What I Saw

  • After 2 days of setting up the whole thing, I saw the web page with the two images started to show up in Google web search when searched for the text used as Alt Attribute.
  • After 4-5 days, when I searched Google with the text used as Alt Attribute, Google results not only showed the page with the images in them but also THE BLOG THAT LINKS TO THE PAGE !

We need to remember that,

  • I used some nonsense text as alt attribute for images which did not have any search result in Google while this was set up.
  • The text used in Alt attribute of the images has not been used anywhere on the blog that links to the image page.

Still the blog showed up when searched for the Alt attribute – How ?

This is only possible, if Google’s algorithm attributes some amount of relevance to a web page for the keywords present on another page that they are linking out to.

I could have actually mentioned the Page URLs and the alt attribute text but the experiment with image optimization is not yet done. So I would better wait to see some results on that before I put those details on the blog.

Read more on SEO at SEO Consultant India Blog

You might also like (related articles):

This article has 8 responses

  • 02.05.2008 8:32 pm Reply

    This test does not necessarily prove your hypothesis.

    Was this test set up with a Blogspot blog? The reason I ask is because if it is a blogspot blog it is part of the blogger network and therefore will be crawled by Google. In essence the domain (blogspot.com) already has inherent links equity even if there are no links coming to your test blog.

    To really test this you would need to purchase a brand new domain to stand on its own and then run the test.

  • 02.05.2008 8:38 pm Reply

    Hi Mark, I sincerely appreciate your thoughts. It is a blogspot blog and though I agree to the fact of the blogspot blog might have some link equity being part of the network, you need to remember the text that was used in alt attribute did not have a single result earlier, which indicates that there is no link strength available on that particular term. Later the blog shows up on that particular term which exists only on one web page in the http://WWW.So I still think it has to be the link to that particular page with the alt text which is the reason for the blog appearing in search result when searched for the alt attribute.

  • 02.05.2008 11:37 pm Reply

    If you can repeat the experiment elsewhere with similar results, you’ll have demonstrated an algorithmic behavior.

    One test really confirms nothing.

    Still, it’s a very interesting observation.

  • 03.05.2008 8:33 am Reply

    Thanks Michael, I definitely plan to run this test again to confirm this findings.

  • Spaggie
    04.05.2008 9:30 am Reply

    This is known as the ‘citation model’ and there is more to it than is described here – go play some more 😉

  • 04.05.2008 10:43 am Reply

    Thanks for your comments Spaggie.A large part of the Google’s algorithm or for that matter, most modern day search engine’s algo is based on Citation model and there is no doubt that this experiment also proves an aspect of the citation model as used by Google.As I mentioned initially the experiment was not initiated to find this type of a result, however, I still bothered to publish it because, these part of the citation model ( benefits due to linking out to an external site) is often overlooked.

  • neo
    05.05.2008 6:36 am Reply

    nice experiment you got … looking forward to see you test on other host and hope to see your SEO result.

  • 04.10.2009 11:04 pm Reply

    Good effort! I am a firm believer in outbound links having a positive impact on PR. I have a personal blog http://asifc.blogspot.com/ that has a PR of 1/10! I know it’s a measly ranking but since I’m not a very active blogger and don’t have many frequent commentors either, I think it must have something to do with my frequent outbound links.

    And of course, whenever I post, I almost always have my own input around those outbound links.