Everyone is. No-one wants to pay for what it takes to get a decent estimate, which incidentally also requires a near-perfect idea of what you're gonna do ahead

Everyone is. No-one wants to pay for what it takes to get a decent estimate, which incidentally also requires a near-perfect idea of what you're gonna do ahead of time which is also usually lacking and which no-one's interested in figuring out and committing to. Look at the space shuttle software development process and how much time they spent nailing down requirements and such so the rest of the process would be highly predictable—IIRC it was like 1/3 of the duration of a project. Ain't no-one in business gonna commit to that. Most software projects are half-cocked and operating on a shoestring. Anything other than "start on this, we'll tell you to stop if we don't like your progress, use something like story points so if we're still doing this in 6 months we'll have an OK-ish way to predict the way the project will go for the next month or two" is delusional. Can it be done? Yes. Here's what it costs. Actual development can start in 3-6 months. You gonna pay for it? Didn't think so. Want us to just start now and see how far we can get in a month or two? Cool.