Startups are essentially machines that build MVPs (minimum viable products) that help answer questions and gradually de-risk the value proposition of the company.

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

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