1 vs 2-tailed test?

[mkdynamic]

Looks like you are using a 1-tailed test to calculate the probability. Is that correct? I see both ways used in different A/B testing tools (Optimzely use a 1-tailed test, I believe).

Curious as to the thinking behind the choice.