My current thought is as follows, operations in business hinges on the new-quality-normalized cycle. This is to say, every new operations maneuver is difficult to qualify without having been run through once. This is the empirical leap of faith that everyone has to take when starting something from (at most near) scratch.
Once the empirical leap of faith has been made, new process began and unrolled, the next phase is qualifying its function in the business as swiftly as possible. At an early-stage, good or bad is almost irrelevant insofar as a judgement is made that corresponds to reinforcement or changes within the process. This is the initial step in the direction of the final component of operations.
Every operational process has to hit the point where normalization occurs in how it functions in the company. Even though chaos is a constant theme in the early-stage start-up space, the best companies don't embody it (too much and especially not at scale). Good companies have engrained, repeatable processes in the core of what they do, whether its in sales, recruitment, product development, etc. The reason for this is, you never want to put yourself or others in a position where you/they have to think more than they execute. Always get more thinking done early.