That makes sense, but is nowhere near universally applicable. For folks working in most startups, talking about different lines of business is non-sensical - th

That makes sense, but is nowhere near universally applicable. For folks working in most startups, talking about different lines of business is non-sensical - there's generally one product. It sounds like you're describing a monolith that actually contained multiple completely independent applications - which I don't think anyone would disagree with being a good case for splitting. In most cases I've seen, the decision to split into microservices usually involves a fairly high amount of dependency between different services to achieve a common business goal (hence the concerns about things like referential integrity)