We are seeing very interesting behavior in our testing of Google +1.
As I mentioned on Rand Fishkin ‘s post a few days ago, we started doing some testing of Google +1.
We tested buying packages of 30, 50, 70 +1’s using Fiverr.
We also tested a system on http://www.plused.net in which you +1 a bunch of sites (in rapid fire from their site) and other people return the favor. We used this to add tens of +1’s to several webpages.
We also tested getting REAL +1’s from a modest group of people who actually like a page.
We did our tests on pages we wanted to move up as well as ones we wanted to move down.
Conclusions so far –
– Google figured out that the rapid fire +1’s and the bought ones were fake and, they actually removed them from the count.
So we saw he numbers go up to say 60, and then the next day they were down to 10 again.
They could be seeing a large group of users with common +1’s and ruling them out. They could also be looking for rapid fire +1’s.
In any case, our conclusion is that Google has at least a basic system for identifying fake +1’s – which is good to know.
We did not see any evidence of fake +1’s causing a page to move in either direction in the SERPs.
We did see a possible connection between the real +1’s and a ~15% bump in traffic for a site receiving 2,500 visitors/day.
That’s it so far – we will keep you posted!