There are software projects that are big. There are software projects that are very big.
And then there are software projects that are too big to fail.
Especially when ignoring the number one rule of failing early and failing cheap projects tend to grow into big scary monsters.
Sometimes these projects work out in the end but what if your project was withdrawn from reality checks for so long that it turns out to be a failure? When you’ve spend lot of time with lots of people costing and spending lots of money and the result in the end does not work?
Does it become too big to fail?
Will you invest more money, time and energy to save it or will you remember the other golden rule never to throw good money after bad?
The longer you wait the more likely it becomes that undoing it will cost a fortune as well. And still sometimes this is the better option rather than to stick to an old formerly good idea that went horribly wrong.
As long as you are able to undo a mistake you should do it no matter what it costs. When it become un-undoable it could even cost your existence..