I came across a curious situation recently. I was reading an article where some text was linked but such that the link text was completely out of context. Here is an example:

as I work my way through the <a ...>best insurance</a> fraud crimes making the news

The linked text, "best insurance" is linked completely out of context as the "best" describes the "insurance fraud crimes". Is there any SEO effect to this? Is there any indication that search engines attempt to analyse context of links as closely as required to detect this?


That link will have value for the page it links to. However, the link won't be very valuable because the page it is on isn't related to the page receiving the link, nor does it rank well for terms the linked to page hopes to rank well for. Basically this link would not be considered a quality link.

The page receiving the link will rank better for the anchor text used in that link. However it probably won't make much of a difference since the linked to page will be lacking many other SEO factors to rank well for that text.


Along with PageRank that everyone talks about, Google also stores lots of information about what it finds on each page it crawls. This includes words used (and count) and words that appear in specific areas (title tag, h1-h6 tags, strong tags, etc). The info the store about a page also includes links out and the anchor text of those links. Basically, they store enough information about each page to get a pretty good idea of what the content on that page is about. This is one of the reasons why AdWords is able to deliver targeted ads, because they know what content is on each individual page.

With that being said, a link from a related site IS worth more than a link from a site with unrelated content. How much more is hard to quantify, but there is a difference. That's not to say I think the type of linking you described in your example is a good idea, mostly because it doesn't seem to add anything for your site visitors...

However, getting links with relevant anchor text from sites with related content is one of the best things, in addition to having quality content, you can do to boost your search rankings


I would highly doubt that google is analyzing whole sentences for indexing purposes (yet). For a search engine to analyze the context they would have to be just one tiny step away from actually understanding the text, they would have to understand that it doesn't fit in the sentence.

search engines have to analyze literally billions of websites, if they were to try and analyze the text it would add millions of dollars worth of processor time. so i doubt it's going to happen in the near future.

  • 1
    Keyword proximity is almost definitely a ranking factor
    – John Conde
    Jul 26 '11 at 0:10
  • I spoke to someone who used to work for Google. He told me that Google translate analysed whole sentences in pages written in different languages that say the same thing to work out it's automatic translation algorithms. So they do analyse whole sentences. Jul 26 '11 at 8:24
  • yes, for translation, and they kinda have to, since the translations would be a whole lot worse if they didn't. but they don't crawl to translate billions of websites a day, they translate a website only when someone 'asks' them to. they do crawl billions of pages a day for indexing, though.
    – xyious
    Jul 26 '11 at 12:12

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.