Startups are essentially machines that build MVPs (minimum viable products) that help answer questions and gradually narrow the company’s value proposition.

The key is that every MVP that builds a business needs to be laser-focused on answering a very specific question. If it does anything more than that, it’s wasted time and effort. In my experience, many startups worry about scaling way too early, wasting resources on something that may never be needed.

This tendency is especially evident in startups founded by individuals who come from engineering disciplines at large, already large-scale companies. But the things you need to do to submit code to Facebook, Netflix, Amazon, or Google don’t apply to early-stage startups in the same way.