Super interesting article. I am not quite sure I agree with everything but I understand what the general message is. The one part I completely agree with is this highlighted bit. It’s something I (and any software/product leader) have to battle all the time. Whenever you are building something you have to balance existing operations, the number of people you have that can work on new features or service scaling, and your backlog. Most of the time, there is no correct answer. You make a choice and you stick with it, and try and do things in relatively small chunks so that you don’t get bogged down in any one thing. But the question I’ve learnt to ask myself now is
If we ship this in n months with this set of features and with these scaling goals will it matter?
It’s not just a way of figuring out what the next step should be (it doesn’t have to be a customer facing feature) and maintaining some level of paranoia. Is it the right approach? Who knows but it seems to work.