A postmortem is honest and useful and always too late. The failure already happened, already cost something. A pre-mortem moves that same honesty to before the work starts, when the failure is still hypothetical and cheap to prevent.
The format is simple. Get the team in a room and say: it's six months from now and this program failed. Not "might fail." Failed. Now write down why. The framing matters. Asking "what could go wrong" gets you polite, hedged answers. Asserting it already failed gives people permission to name the thing they were quietly worried about.
You'll get a list of failure modes in about twenty minutes that no risk template would have surfaced, because they come from the people closest to the work. Then you sort them: which are likely, which are fatal, which are both. The both pile is your real risk register.
The whole thing takes an hour. I've never run one that didn't change the plan. Optimism bias is real and it's strongest at kickoff, exactly when the stakes of a wrong plan are highest. A pre-mortem is the cheapest counterweight I know.