> This is why I think startups or small groups of people are still able to outcompete bigger incumbents. For sure. The overhead for communication is much lower

> This is why I think startups or small groups of people are still able to outcompete bigger incumbents. For sure. The overhead for communication is much lower. It’s possible to emulate this to some extent at a large company (I worked at AWS previously) by keeping teams small (6-10 people) while keeping team scope large (full lifecycle of an entire product family or vertical). For this to be successful I believe you need to give the team autonomy to decide what product and features to build, and room to fail without repercussions to allow for experimentation and calculated risk taking.