SEO theories I can’t prove


This is a post I started writing ages ago and didn’t finish. This week people have been talking about Google’s latest revelation that they’re using Tweets and Facebook likes as part of their organic search algorithm. If you read my post on building links to ecommerce product pages you’ll know that this is something I’ve been talking about for quite a while now. Hopefully if you read that post and believed me when I said this you’ll already have been working on ways to get deep links from Tweets and Likes and will be ahead of the curve with this already…

One of the ideas I’ve had in my head for a long time and I think seems to work, although I can’t prove it in any real way, is that Google is using a new (or probably not so new) scoring system around popularity signals. This is seperate to trust or relevance and doesn’t necessarily rely on good old fashioned followed text links so anything from tweets to facebook likes to unlinked mentions to social bookmarks to Analytics data to search query data and beyond is all contributing to some rankings, although conventional SEO wisdom may say otherwise.

(I think thats the first time I’ve ever quoted myself, it feels good!)

This made me think that I’ve got quite a few other SEO theories which I’m never going to get around to proving and which Google will probably never come out and validate but which people might be interested in hearing anyway. Below is a list of stuff which I believe to be true and which influences the way I do link building, but which is nothing more than theory.

  1. PageRank is still really important and the best single metric to look at when grading link prospects
  2. Recipricol links are one of the best ways to build links for small businesses. Most link building tactics go full circle and start working again if you stick at them long enough.
  3. Domain authority is a myth and doesn’t work in the way most people think
  4. Anchor text isn’t as important as you think
  5. Internal links are more important than external links
  6. Image link alt text is as, if not more effective than text link anchor text
  7. Social bookmarking is almost completely useless as a link building tactic (I’ve only come round to this point of view recently)
  8. Google doesn’t really mind paid links (they just don’t like link manipulation in general)
  9. Duplicate content is less of an issue than you think
  10. Links from twitter and facebook can have a massive impact on rankings
  11. The sandbox effect is more prevalent than ever
  12. Bad links can really mess up good sites if they’re new and untrusted
  13. Places search will become the new playground for blackhat SEO’s as its so easy to spam at the moment
  14. Product and place reviews will help you to rank and Google will never be able to filter fake reviews effectively
  15. Even low quality directories have some value if you’re smart with how you use them
  16. Server IP location is nothing to be concerned about for most people
  17. Links from irrelevant sites aren’t devalued in the way many people think
  18. Google is experimenting with using seasonal search trends as part of its ranking algorithm and rankings in some industries are likely to start fluctuating more by season
  19. Your link building effort should increase as your rankings improve, which means you should be doing most of your link building to pages/ keywords where you’re already number 1
  20. Position of links on a page is a very important ranking factor
  21. Improving page load speed is the highest ROI activity most ecommerce sites can undertake (and is a more important ranking factor than most people think)
  22. Google aren’t nearly as concerned with what SEO’s are getting up to as most of us think!

This is just a collection of thoughts but I’m happy to answer any questions on these ideas. I want to blog more next year so I’ll go into more depth on some of these theories then.