> "Since we made those changes, we have noticed the ratio of people who try our presentation demo now is up from an embarrassingly low 1% to a much healthier 28%."

I think you've made positive changes, but don't forget that your organic audience is much different than the burst traffic from a link aggregator like Hacker News.

Not that your conclusions are necessarily wrong, but it's easy to make false correlations when more than one variable changes.