Jameson, CEO, here. We haven't done A/B testing against the previous site, but we closely watch our metrics and have seen a lift since the new product went out. We were confident in the new app, and would have rolled back if necessary (it was out for a week before we announced it), but all signs were good. Actually setting it up to do a proper A/B test would have taken a lot of engineering time, and we judged it to not be worth it at this point. A lot of that has to do with our goals which are more about expanding the platform than optimizing the "launching soon" feature set.