There is a particular kind of meeting where everyone agrees, everyone leaves, and everyone has understood something slightly different. I have been in that meeting. I have caused that meeting.
The fix is almost never a better document afterwards. It is a better question during.
Closed questions get you agreement, not understanding
"So you want the approval to happen before submission, right?" invites a yes. And people will say yes, partly to be helpful and partly because the version in your head sounds close enough to the version in theirs.
"Can you walk me through what happens today, from the moment the request comes in?" gets you something different. It gets you the workaround nobody mentioned, the exception that happens every month, the person who has to be called before anything moves.
Ask about the last time, not about the general case
When people describe a process in the abstract, they describe the version that is supposed to happen. When they describe the last time they actually did it, you get the real one. This has been one of the most useful adjustments I have made.
The questions I keep coming back to
- Who is this actually for, and what are they trying to get done?
- What happens today, without this?
- What breaks if we get this wrong?
- What would make you say this worked?
- Is there a case where this rule shouldn't apply?
- Who else sees this differently?
None of these are clever. That is somewhat the point. Most of the value comes from asking them consistently rather than from asking anything unusual.
Not knowing out loud
The hardest part of asking good questions is being willing to look like you do not already know. Early on I would nod through explanations and try to reconstruct them later. It is much faster to say "I've lost the thread, can you take that part again?" in the moment.
The cost of asking is a few seconds of feeling slow. The cost of not asking shows up much later, and it is paid by someone else.