There's possibly a good analog in data. A while ago it seemed very popular to talk about "schemaless" data systems, and I heard the perfect commentary: "If you

There's possibly a good analog in data. A while ago it seemed very popular to talk about "schemaless" data systems, and I heard the perfect commentary: "If you think your data is schemaless, what that really means is you don't know what the schema is, and that's bad." What people probably meant by schemaless: It's very easy for things to have a slightly different schema than they used to if that's needed. Things that don't all conform to the same schema can still be handled well. And it should be understood that sometimes you do need a better defined schema. So my point? Don't focus on being 'leaderless' or 'structureless'. Focus on being flexible enough to allow new substructures in the company as needed. Make sure there's a system in place that allows people to coordinate with groups that don't have the same structure as them. And recognize when you do need a bit of structure to keep scaling.