|Can Google Manipulate Traffic?|
I just saw a post in here that talks about Google "releasing" (poor quality) traffic and in the past there have been posts about Google "throttling" traffic. The implication is clearly that Google can physically control traffic flow to sites.
Am I the only skeptic in here? How would anyone come to the conclusion that Google can, or would, manipulate traffic flow other than by algo tweaks that bring about a change in the SERP's?
People who come to that conclusion are analyzing the traffic they get. They see a sudden change in the normal variations, up and down, and all of a sudden the traffic graph is almost perfectly level. That's what those people call throttling.
And in recent times, some webmasters see the same raw numbers of traffic producing a drastically lower level of conversions. Then when they dig deeper into their stats they see things like a major shift in the countries of origin, or sometimes time of day.
You're right Google "controlling" traffic is not a proven thing - it's a theoretical stab at explaining these odd changes in website stats for traffic from Google search. And if Google is doing this intentionally, then you could call whatever they are doing part of the algorithm.
I personally think there is more than "one ranking algorithm" in effect most of the time. If that's true, then the way the various sets of rankings switch in and out would also be done by an algorithm, maybe done for some kind of statistical testing purposes.
And my idea also is just another unproven theory. However, I have seen enough examples of strange effects in website analytics (not just Google Analytics, either) to feel that there is something going on, even if we don't have a good handle on exactly what it is. It could be an accidental side effect of something else, or it could be some intentional change... we just don't know for certain.
But we do know that two years ago only a few webmasters were reporting anything like this, and today there seem to be a lot more.
I am absolutely certain Google is doing this now more than ever; particularly in areas they want to ensure traffic goes to their own properties - such as local search.
One of our sites is in that area, and we've seen really clear examples where - for example a place we rank well for is covered in a TV show and we receive a couple of hundred extra organic visits from Google for the relevant page - later that day, organic Google traffic across the board will fall off to "make up for it". Traffic to our site is (or at least was until Panda came along) pretty predictable and these anomalies occur too often for it to be coincidence.
There is also very clear a "glass ceiling" in place since the start of the Panda roll out (which coincided with Google making their Places listings a whole lot more prominent). By this I mean, looking at our organic traffic from Google on an hour-by-hour basis, it follows a smooth curve - peeking at a certain times in the day. On occasions, again particularly when some of our content is attracting extra interest, this sooth curve is truncated at the top with a few hours receiving almost exactly the same amount of organic traffic from Google. Looking back over years of data, this simply isn't how people search.
I believe that personalised results are an excuse to shape traffic to other websites, particularly those that compete in areas Google wish to push their own properties. At the moment if I carry out the same local search from different browsers, on different IPs, I get different results - both in terms of who ranks organically (the top 3-5 results swap around a lot) and how how prominent the Google Places 'results' are and what form they take. I fail to see how this kind of inconsistency provides benefit to users, or is a result of personalisation (clean browser, not logged in) -- but I can see how it would be used to shape traffic to websites.