Binning everything apart fom the twitter RT button above..

 But wouldn't that mean that if A isn't there then B will automatically get higher figures rather than the balance between the two.

As someone has said above is it causing a problem by having the various options there?  Is there a post somewhere of how to use the forum at all and what does what/goes where?  Perhaps one that's stickied that can be amended by TPB from time to time would be helpful.

David - Not sure you understand A/B testing.

In essence:
Month 1 - Run with the element (for example 'tweet this' button) in place.
Month 2 - Remove the element.
At the end of month 2 compare results.

Lee

I'm reading it that you're saying Month 1 all the buttons are there and then Month 2 remove one of the buttons.  Or am I wrong and you're saying Month 2 remove ALL the buttons?  If people can't exercise their preferred social network button then they will automatically default to one of the others or not at all so 50/50 whether sharing would be lost.

No. I mean, for example:
M1 - Have the twitter button in place
M2 - Remove it
Test and repeat with the rest or other options.

This is just split testing at it's most basic.

Having not really noticed those buttons before I have just hovered the mouse pointer over each of the buttons expecting a mini help. Only hovering over the "Email" button achieves anything and that's a pop up asking me where to email. At the bottom of that pop up there's a further choice apparently allowing the same "E", "T", "F", "in" choices plus Y!

It may help people decide to use those buttons if there were a mini-help pop up available to say what they are for.

Lee

I was correct in my understanding but perhaps didn't expand sufficiently.

It is my belief that if you remove any of the options if somebody is wanting to share a post then they will select from the other options available because it's the best option they have available to them rather than being able to utilise the option of their choice or worse case scenario of not sharing the post at all.

Ah, apologies then, David.

I kind of agree with you but also that A/B testing is a more complicated science. My example above was just a simplistic example.
That said, in the 'button scenario' you'd possibly find more people using twitter than G+, generally, depending on factors like poster, topic, longevity, relevance ... so the results would be kind of skewed anyway.
Normally you'd do things like test with/ out, change the position, size the tweet button, with/ out G+ next to it/ on the left/ right ... the list goes on.
 

 No worries as easily done :)