Building LinksAnother test ‘result’ – Do links high in code matter more than links at the bottom of a page (as Google reads it)?

Background

There’s been a lot of complaining from webmasters about Google’s recent SERPS update, which saw a lot of webmasters flock to forums to complain about ‘poor quality’ sites ranking after the bug ridden update (confirmed by Google).

Matt Cutts of Google made a rare posts at WebmasterWorld explaining that this was a bug:

I think this was a short-term issue and things should be back to normal pretty soon (if not already).

Matt clarified later in the thread:

I don’t consider those rankings indicative of anything coming in the future. Some data went into the index without all of our quality signals incorporated, and it should be mostly back to normal and continuing to get back to normal over the course of the day.

Yes, this was a bug as Matt said in reply to a question as to whether or not this was a “bug” or a “test.” Matt said, “nope, it wasn’t a test.” But was it? ;)

Google Test Results

As you might know, I test a lot of stuff (see below), and recently, I had a test running trying to determine which links where most valuable on a page. For instance, placing 3 exactly identical text links on one page to 3 different pages. The only links these pages had.

The pages where not identical either, although they a had very similar content and nature, they were enough to feature in the SERPS. Caveat – they were probably supplemental pages and the word had no competition in the SERPS.

I have had this seo test running for months. The results showed clearly, when all is equal (it never is by the way in real life) Google will count links higher in the code as more important when ranking pages in the results pages.

Example – Just visualise removing all the copy on the page, so Google just reads the links as a list of links (important to note these where internal links on another site):

  1. Link (First link in the code carries more weight)
  2. Link (This link was second)
  3. Link (This link was least effective)

It did not matter how often I linked to the pages from that one page – Google just seemed to count by order of how it discovered them as it scanned down the code (it has nothing else to go on).

Interestingly, during the update, which a lot of webmasters where complaining about, this test ranking was mucked up for the first time I had been monitoring it. Now it seems Google has admitted a mistake (fair enough) and are now resolving it (we don’t know how obviously), but it seems as though the test results have returned back to what they where before this update.

Perhaps Google had been attempting to introduce a fundamental change in the algo how they treat links (perhaps especially internal links) and it may have had unforeseen detrimental effects in resulting serps?

Summary

Anyway, it seems links higher up in the text content mean more to Google again than links it finds below the first link (in this admittedly small-scale test).

BIG BIG note – it seemed to take some time for the results to settle, i.e. this was not an immediate effect. It seemed to settle over a long time.

PS – I hope I am not a seo that is making people dumb.

If anybody wants to see the actual tests as I have them running, please PM me.

Recent Hobo Google SEO tests you might find interesting

  1. Google Ranking – How many words in links are a benefit?
  2. Which link counts when duplicate links are on a page?
  3. Does W3C compliance boost your Google ranking?
  4. How many words will Google rank of a page title?

Add – These tests are never published in an effort to show you how to screw Google, because I don’t think that is in anybody’s interests – you should stay within Google and every other search engine guidelines, more me just sharing test results I use to help me when developing internal link structures.

It would be great to hear others test results especially:

  • If you think the same,
  • or you have evidence Google works another way in this matter, because it might very well do site to site.

I think about this sort of thing when I am building sites, and perhaps you might too. The more people test, the more we can argue :)

PPS – Test everything for yourselves.

Did you know when you link to a Hobo SEO post we have search engine friendly links back to your site if approved? Our comments are also search engine friendly you know (once you've commented on a few posts)! Do you need any more encouragement to get involved in the conversation ;)