You have a greenfield development project with a very well understood set of business needs and requirements, and you have the budget (I'm guessing) to run some small-scale experiments without the usual sort of time pressures; to a certain kind of developer (hi!) this is super attractive.
I would spend a few weeks getting to know big chunks of the core use cases, and then architect a system designed from the beginning to handle those needs, and the prototype a bunch of that system over a couple of months.
Although these time estimates are totally armchair bullshit, what I'd actually do is timebox both steps to something like "two weeks" and "two months" - that'll help mitigate scope creep that'll come from not having the usual sort of time pressure. It's not about getting it done, it's about finding out what it would take to get it done, and seeing how much can be accomplished in a set amount of time is a good way to do that.
The key bit in all cases is (1) hiring good people (2) getting out of their way except for (3) keeping them on track. (One trick for #3 is to have them propose plans, and then always ask "okay but is there a simpler way to do it?", 2-3 times, kinda like 5-why's).
For #1, I'd look at talks people give at conferences, and then either try to hire those people, or ask those people for referrals.
For #2, that's your executive-level corporate culture. Unfortunately, from what I've seen, you either already have it or you don't, and that's probably not something you can change because if you don't have it, you also don't have the psychological safety necessary to find out that you don't have it - although, you can look for where you've got high employee churn, and that's where your problems are.
For #3, I'd use a combination of what I call the "wine trick" with "ELI5". The "wine trick" is that you don't actually need to know anything about wine to find out if someone else does: just get them talking about the details of their special interest, and if they can, they know a lot about their subject. Combine that with "Explain It Like I'm Five" to find out if they're actually just bullshitting, and because the other way to find out if someone knows a lot about something is if they can teach it. (Plus, they'd need those communication skills during the rest of this process anyway).
You have a greenfield development project with a very well understood set of business needs and requirements, and you have the budget (I'm guessing) to run some small-scale experiments without the usual sort of time pressures; to a certain kind of developer (hi!) this is super attractive.
I would spend a few weeks getting to know big chunks of the core use cases, and then architect a system designed from the beginning to handle those needs, and the prototype a bunch of that system over a couple of months.
Although these time estimates are totally armchair bullshit, what I'd actually do is timebox both steps to something like "two weeks" and "two months" - that'll help mitigate scope creep that'll come from not having the usual sort of time pressure. It's not about getting it done, it's about finding out what it would take to get it done, and seeing how much can be accomplished in a set amount of time is a good way to do that.
The key bit in all cases is (1) hiring good people (2) getting out of their way except for (3) keeping them on track. (One trick for #3 is to have them propose plans, and then always ask "okay but is there a simpler way to do it?", 2-3 times, kinda like 5-why's).
For #1, I'd look at talks people give at conferences, and then either try to hire those people, or ask those people for referrals.
For #2, that's your executive-level corporate culture. Unfortunately, from what I've seen, you either already have it or you don't, and that's probably not something you can change because if you don't have it, you also don't have the psychological safety necessary to find out that you don't have it - although, you can look for where you've got high employee churn, and that's where your problems are.
For #3, I'd use a combination of what I call the "wine trick" with "ELI5". The "wine trick" is that you don't actually need to know anything about wine to find out if someone else does: just get them talking about the details of their special interest, and if they can, they know a lot about their subject. Combine that with "Explain It Like I'm Five" to find out if they're actually just bullshitting, and because the other way to find out if someone knows a lot about something is if they can teach it. (Plus, they'd need those communication skills during the rest of this process anyway).