They could also lift your rating - but still with the randomiser - through measuring click through and return rates. If when your luck comes in and you are page 1 and you then get a higher proportion of clicks then you must appear to the human eye to be a good starting point. If they then come back in a few seconds maybe that tells them something else also.
Don't they already do this?
Randomiser? I very much doubt that there is or ever will be a random element as a designed-in factor for google's SERPs algorithm. When results might appear random or fluctuating it could be due to load balancing, whereby google decides (perhaps with some element of randomness) which server (data centre) to process your search query, this can change by the second, so the same query may not return the same results if different servers are used, and those data centres are out of synch, which is more noticeable during an update (Dewey).
-- EDIT by Paul, 26/05/08 23:10
I wrote the following paragraph a few hours ago stating what I believed, at that time to be true. Since then Duane has posted to this thread some new information, which has made me question my initial belief, and so I will look into this further and report back. Please take the following paragraph with a pinch of salt, as it could be wrong, and read the thread in its entirety for the full picture.
-- EDIT
As for google monitoring clicks, perhaps it is all these big brother news stories we get, or people mistaking one area of google with another, but I do think that a lot of people believe, or would easily accept the belief, that google is monitoring what results everyone clicks on. This is simply not true. In its normal use, google does not monitor which natural results are clicked on, simply because these are direct links to the site in the search results. Google does monitor which google ads results are clicked on. In normal use for natural searches, out of google, yahoo and MSN/Live, only yahoo monitor clicks in their natural results (check the urls of the search results to see this).
I mention normal use, since there are other ways google can monitor this kind of thing, for example:
1. If the user has the google toolbar installed on their browser, with the google pagerank option active. But not everybody has this, and the default installation (the latest one at least) does not have the google pagerank option active. If you activate the pagerank option you are presented with a warning page saying that there are privacy implications of you using this.
2. If the website the user visits from google has other google software on it, whether it is google analytics, google ads, google adwords tracker code, etc
So whilst it is possible for google to monitor which natural results have been clicked in certain cases, google cannot monitor for the majority of cases of normal internet use. In these special cases where google can monitor, whilst google might be saving a lot of monitoring data, it doesn't necessarily mean that it is actively doing anything with it (at the moment) - just like supermarkets with the masses of personal information they capture when customers use club/loyalty/points cards (what people buy).
I can give a couple of examples where even if google could capture this information, it would be meaningless anyway:
1. What constitutes a useful site and what constitutes a useless site? If google can monitor what search results get clicked on and monitor whether you visit a number of pages within that site, what meaning can we get out of how many pages on the site were viewed. A useless site might have one page view (you know straight away the site is useless), or several page views (you browse the site and then realise it is useless). Likewise a very useful site could have exactly the same pattern, one page view because exactly what you are looking for is on that one page, or several page views because it is interesting.
2. When you click on a few results how could google work out that one is more appropriate than another? Time? If you click on result 1, then 5 seconds later you click on result 2, the 5 seconds later you click on result 3 that could tell google that results 1 and 2 were no good and result 3 was better? But what about people who multi-task, and want to compare different sites (e.g. prices), and simply open up all 3 results in different tabs on their browser, or 3 different windows for later viewing?
There is one area where I think monitoring of clicks might help (which Page mentions), but I doubt that google are doing this (yet)...
For a given search term if google can monitor that certain results that rank highly are never clicked on for certain searches (e.g. pet shop boys, the pop band, sites when you are searching for
pet shop), because although google's computers and algorithms think they are relevant, the human user knows they are not relevant because of the result listing and description, then monitoring in this situation could eventually downrank the odd result. But this system would be open to abuse (just like competitor adword click abuse), and error prone too (just because a listing looks irrelevant it still might be very relevant).
Apologies for dragging this thread further off-topic. To get back to the original poster...
I wonder if others are having issues with Google since the recent changes to the algorithms. I have been swaying between 3 and 70 for business coach and my SEO has similar problems with virtually all the sites he is working on.
Yes, we are noticing some fluctuations on a couple of sites over the past few weeks, sites which were page 1 or 2 of google but then drop completely out of the top 20 pages, and then come back (one hasn't come back yet). It could be the google dewey update, as Ray mentions, or it could be something else, as Scott mentions, (although in these cases it is not due to UKSBF since the sites in question have no inbound links from UKSBF)
not even in the top 100
