When should a startup actually go with a monolith?

Everyone talks about microservices and distributed systems, especially when scaling. But if you're a new startup, with not many resources and you dont even know if your idea will work, starting with a monolith can make things a lot simpler at first and speed up how fast you build stuff. So, when would you really push for a monolith, even if you know you'll probably have to break it up later? What are the main signs that a monolith is the smart play for a startup?