Without going too much into it a lot of the performance issues stems from the current public API and how our data model is set up. Basically it hasn’t been scaling well. For a while we were tossing resources at it.
We roadmapped some quarterly goals to address this issue and started to work towards them since the beginning of the year. So far we have been meeting each one (although many are back-end stuff you all probably won’t notice immediately).
I’ve made a post earlier in the day as an aside on a different topic.