Before the meeting, everyone who has a stake in what's being discussed contributes to a memo outlining their thoughts on it. Then everyone reads it ahead of the meeting.
The meeting is then just to hash out what's in the doc and agree on which already proposed solution will be adopted.
In the first org, there was a lot of pressure that any such memos had to be highly researched and data driven (also borrowed from Amazon, I imagine). In reality what this meant is that drafting a memo even for the smallest decisions or ideas became a large laborious effort, which nobody wanted to do and/or didn't have time to do it. The result was that everyone procrastinated on decision making, if they even participated in it in the first place, and I knew a fair share of colleagues who didn't even bother to pitch their ideas because of hating the memo system.
In the second org, it was the opposite problem: in an attempt to make memos "painless", there was much less pressure on making the memos "high quality". The result: every "memo" was a hastily scrawled together draft of random notes that didn't make much sense, and we had to spend most time in meetings going over the "memo" to decipher it and have the author explain it anyway.
Meetings where people are there just to listen in, or for FYI purposes are utterly useless.
Also, the fewer attendees the better, in my experience. 3-5 is a good size. 1-2 "higher ups" and 1-3 with "boots on the ground". More than that and it risks devolving into chit-chat and speculation instead of bullet-pointed queries and actions.
2-4 person meetings are great because everyone has a chance to interact and ask questions.
I've also personally noticed an interesting subtle difference: in-person there seems to be a tendency to contribute if someone is present in the meeting and hasn't said anything. With zoom the general tendency seems to be the opposite (the longer someone is on mute, the more likely they are to stay on mute).
Anyone else notice things like this?