Did I make the right decision?
This question usually arrives late, either after something went badly or during the long stretch where you can't yet tell. Either way it's being asked about a decision you can no longer change, which means the useful version of it is not "was I right" but "was that a reasonable way to decide, and what do I do with the answer".
The two are separate, and telling them apart is most of the work.
A bad outcome is not a bad decision
You can decide well and lose. Every choice worth journalling is made under uncertainty, which means a good process still hands you a distribution of outcomes and you get exactly one of them. Judging the decision by the one you got is a documented error rather than a personal failing: the psychologists Jonathan Baron and John Hershey ran the experiment in 1988, giving people identical descriptions of a surgeon's decision to operate and varying only whether the patient survived. When the patient died, the same decision, made on the same information, was rated worse. Physicians judging colleagues did it too.
Annie Duke, who left a cognitive-psychology doctorate to play poker professionally, calls this resulting, and it runs both directions. The reckless call that happened to pay is the more dangerous half, because nothing about a good outcome prompts you to examine it.
So if the honest answer to "did it turn out well" is no, that isn't yet an answer to the question in the title, and if the answer is yes, you still haven't got one.
What you can actually check
Five questions, all answerable now, none of which depend on how things went:
Could the prediction have been marked wrong? Try to state what you expected to happen, in the words you'd have used at the time. If the best you can produce is "I thought it would work out", there was no prediction to be right or wrong about, and there's nothing here to grade.
Did you name what would change your mind? A decision with no stated disconfirming evidence tends to be one you'd already made. This is the single strongest signal in the list, and it's the one people most often find missing.
Did you consider the rejected option at its strongest? Or did you compare the best version of what you wanted against the weakest version of what you didn't? Reconstruct the case for the other side now and see whether it's better than the one you dismissed.
What did you know then that you're now discounting? Something you were told, and set aside, has since turned out not to matter — or to matter enormously. Both are worth finding, and the second one is the only genuine mistake among them, because it was knowable at the time.
What state were you in? Tired, rushed, flattered, cornered, or on a clear morning with time. This isn't an excuse and it isn't a verdict; it's a variable, and across enough decisions it turns out to be one of the more predictive ones.
Answer those and you'll have a fair read on the process. One caveat, and it's a large one: every answer above is coming out of a memory that already knows the outcome, so treat them as the best available evidence rather than as testimony.
The part you can't get back
There's a limit on how honestly this can be done after the fact, and it isn't about effort or willingness.
You don't store your old estimate and hand it back on request. You rebuild it, and the rebuild draws on everything you know now, including the result. Told that a project failed, people don't merely conclude that it failed; they recall having been less confident than they were, and they can't tell the difference between the recollection and the original. Hindsight bias doesn't weaken when you're warned about it, and it falls hardest on the people conscientious enough to review their own calls.
Which is why the answer to "did I make the right decision" gets less reliable with time rather than more, and why no amount of careful reflection recovers what a single written line would have preserved.
The decision in front of you
The one behind you will give up what it's going to give up in about ten minutes of the questions above. The one in front of you is where this is worth anything.
If it's reversible, it deserves less agonising than it's getting; find out what undoing it would actually cost, because the number is usually lower than the feeling. If it isn't reversible, spend five minutes on the entry that makes the next version of this question answerable: what you expect to happen, stated so a stranger could later mark it right or wrong, how confident you are as a percentage, the two or three reasons, and what would change your mind. Then put a date in the calendar for when you'll know.
Do that a dozen times and this question stops being something you ask a memory. The template has the fields, and the method essay explains why each one is there.
Full disclosure: I make Reckon, an iOS decision journal built around this exact gap — it holds the prediction and the confidence you logged at the time, brings the decision back on the date you set, and asks again a month or two after it resolves, when you can judge the call without the outcome still sitting on top of it. One-time, no subscription. Paper does the same job if you'll keep it up.