Robert_Charlton - 9:26 am on Nov 7, 2012 (gmt 0)
ichthyous - Your posts raise a bunch of thoughts.
First, I'd discount the idea that reactivating AdWords would affect your organic ranking. I myself am convinced there's no cause and effect, at least not in the sense that having an account caused Google to treat you differently.
A symbiosis between organic clicks and AdWord clicks has been noticed for many years, though, so conceivably that might have some effect on traffic and user behavior in a way that's now measured by the algo and boosts rankings... but I think this would have been more widely noted by metrics people if it were so.
Reducing 1,200 dupe titles to 22, however, would likely have a direct effect on rankings, and a month (plus the recent update) could allow enough time for the algo to have factored those fixes into your rankings.
You make a fascinating observation about different rankings on different browsers. If I understand you correctly, you're suggesting that perhaps having a mobile unfriendly site might have caused your rankings to drop on mobile browsers.
That resonates a lot with a post I made in October, in this thread in our Local forum, about whether a mobile unfriendly site could hurt your rankings....
43 Percent Of Total Google Search Queries Are Local
...And in our last month's Google serps update thread, "a huge shift" towards mobile was discussed around Sept 20, 2012, with many confirmations that the traffic was not just local traffic from mobile phones, but also B2B on "mobile ecommerce on tablets, not phones".
Google Updates and SERP Changes - September 2012
I don't think I've ever noticed so many such reports in one update thread. Sites that didn't have mobile versions were reporting lost traffic, and a high abandonment rate that could hurt them in Panda.
What you say about "IE9/PC ...still on page three", though, kind of muddies the water a bit regarding the browser/device correlation... but what really intrigues me is the idea that Google might be "personalizing" site rankings to fit their friendliness to a searcher's browser. This is thinking out loud, and I haven't thought through the mechanisms by which Google might do this.
I'd want more precise data before making any assumptions. As you note, it is possible in Google Analytics to filter your results by browser, etc, and you can pin down the correlation more tightly than you've reported it. The idea is worth following up.