Yes, early stage startups must go super narrow and deep, and solve one problem only. Absolutely true.
But sometimes I see founders misunderstanding this as skimping on critical features, without which the product cannot deliver its proper value.
Consider Uber v1. It was a narrow service – all it did was hailing limousines in SF. But in order to make that one thing work, the app had to have all stacks ready. If one stack was missing, the whole thing wasn’t going to work. Imagine every stack of Uber working except for the payment stack or the driver dispatching stack.
Startups are not just solving one problem, they need to solve one problem really, really well. Think about the minimum stacks and components required to deliver a magical, 10x experience to your users, when it comes to that one problem you’re tackling. You should focus, which means defining what not to do – but at the same time you don’t want to skimp on the minimum stacks and components needed to deliver a 10x experience.