05-02-2013, 06:44 AM
Hi - there are loads of threads about ways to rank on Google but I have seen very little information on the workings of Google: if we don't understand it, how can we hope to master it, except by trial and error?
I was prompted to post this by seeing some odd behavior by Google that got me thinking about the fundamentals. It happened like this: I was trying to see where one of my sites was ranking for a specific keyword, and decided to see what variations there were using different proxies. I was very surprised to see that there were large variations, and the variations were very repeatable. So the answer Google gives is dependent on what it thinks is the browser's IP address.
What this suggests to me is that Google, as a massively parallel indexing and computing platform perhaps has delays running through it - there isn't one answer to the question "where is my site ranking for a given keyword" because, in reality, there is no single Google doing the calculation. It's as if Google has multiple personality disorder - at least, that's my thought.
This may account for the "Google dance" - seemingly chaotic behavior as a site shoots up and down the ranking. It reminds me of optical interference where multiple paths through a system serve to increase or decrease the sum total of a signal depending on each path length. Perhaps what is happening with the Google dance is different rankings being calculated in different parts of the Google brain, adding up to an unstable total that takes time to settle down. I haven't expressed that very well, but maybe you can catch the gist of it.
Why does any of this matter? Well, understanding the mechanisms by which Google ranks a site could lead to better methods of improving rankings, and also help us discern the truly useful ideas from the vast number of BS techniques out there being pushed by unscrupulous snake-oil salesmen and women.
I'm only proposing some thoughts - they aren't even a theory - but it would be interesting to hear what others think and whether anyone has more insight on how Google actually performs its magic.
I was prompted to post this by seeing some odd behavior by Google that got me thinking about the fundamentals. It happened like this: I was trying to see where one of my sites was ranking for a specific keyword, and decided to see what variations there were using different proxies. I was very surprised to see that there were large variations, and the variations were very repeatable. So the answer Google gives is dependent on what it thinks is the browser's IP address.
What this suggests to me is that Google, as a massively parallel indexing and computing platform perhaps has delays running through it - there isn't one answer to the question "where is my site ranking for a given keyword" because, in reality, there is no single Google doing the calculation. It's as if Google has multiple personality disorder - at least, that's my thought.
This may account for the "Google dance" - seemingly chaotic behavior as a site shoots up and down the ranking. It reminds me of optical interference where multiple paths through a system serve to increase or decrease the sum total of a signal depending on each path length. Perhaps what is happening with the Google dance is different rankings being calculated in different parts of the Google brain, adding up to an unstable total that takes time to settle down. I haven't expressed that very well, but maybe you can catch the gist of it.
Why does any of this matter? Well, understanding the mechanisms by which Google ranks a site could lead to better methods of improving rankings, and also help us discern the truly useful ideas from the vast number of BS techniques out there being pushed by unscrupulous snake-oil salesmen and women.
I'm only proposing some thoughts - they aren't even a theory - but it would be interesting to hear what others think and whether anyone has more insight on how Google actually performs its magic.