When Something Works
With startups, effort is everything, but you have to be putting your effort in the right place. A reminder to myself: when something works, do it more. When something doesn't work, don't do it again or change the process entirely.
When something works really well, figure out how to multiply that activity by 100 or 1,000. Find an activity that converts into a positive outcome then create a repeatable process. If you have the capability to do that activity many times over, that's when you can achieve scale.
The question of what works is interesting (that's 1. market testing/validation), but the question of how you can multiply that positive activity many times over is exciting (2. system thinking/scale). You want to design systems that compress the decision-making process of steps one and two as much as possible.
Ideally, you want to discover what works quickly, and then quickly determine how to multiply what works.