I believe you're going about this in the wrong way.
I'm sure you can find a list on the internet somewhere of "important metrics to track", but what you should really care about is "what am I trying to learn?" In general, we:
Build → Measure → Learn
But to properly execute a single cycle of the "build, measure, learn" feedback loop, we have to go backwards. We must know the questions we are trying to answer (what we're trying to learn), to better understand the metrics we should track, and the metrics we're trying to track heavily influence what it is we're going to build.
So ask yourself:
- What are my assumptions?
- If I change feature X, I expect more users to sign up.
- What metrics do I need to track in order to prove my hypothesis?