FYI that's not a bell curve. It looks much more lognormal or Beta to me (or even Weibull or Gamma could fit it, if you didn't care about how meaningful the par

FYI that's not a bell curve. It looks much more lognormal or Beta to me (or even Weibull or Gamma could fit it, if you didn't care about how meaningful the parameters were, or wanted to get creative in interpreting them). If you go with Beta, Weibull, or Gamma distributions, mean and standard deviation are the wrong vocabulary to use. Obviously you can't tell just by looking at the plot, but it's an easy test to show definitively if it's normally distributed or not. The difference here is meaningful, depending on how much you want to read into the distribution of open rates.