36 conversions/3784 visits sounds like random fluctuations to me. Also, was it video+text, or video instead of text?
Re:What's that?
by worldvoyageur on Thursday March 22, 2012 @08:31AM
(#3739397)
Another way to say conversion went from 0.5% to 1.3% is to say visitors were over 2.5 times more likely to convert if there was video than if there wasn't. In those terms, random fluctuations seem less likely as the reason for the difference.
FWIW:
> chisq.test(matrix(c(10,3784/2,26,3784/2),2,2))
Pearson's Chi-squared test with Yates' continuity correction
... p-value = 0.0129
Granted the contents in the video are not the same as the explanation text is, so might be attributable just to the difference in content. Or possibly just random chance (36 conversions 3784 visits).