There is a problem with asynchronous text-based communication that doesn't happen with synchronous voice-based communication: I often write (or feel the need to

There is a problem with asynchronous text-based communication that doesn't happen with synchronous voice-based communication: I often write (or feel the need to write) text like this: > I'm not sure what you meant by "blargh". If you meant foo then [long explanation, probably with bullet points]. If you meant bar then [a different long explanation]. This isn't an efficient use of either of our time. Also, if the explanation was already a bit intimidating (in terms of amount of content) then double that is even worse, even if half of it is actually irrelevant. The alternative is that I ask what they meant by "blargh" but often the response doesn't get the heart of the difference, probably because the difference is subtle enough they didn't realise there was one in the first place. So it takes a few messages to sort out, by which point the whole conversation has dragged on for a few extra hours precisely because the communication is asynchronous so messages aren't replied to instantly by either of us, which is distracting us (or me, at least!) for that long. In a conversation where there's likely to be some ambiguities in what you're saying, it's just much faster and less wasteful to iterate your joint understanding with a synchronous conversation. I often have highly technical conversations that are like this, but it's probably true even of lots of non-technical conversations. Of course what I've said is mostly true of one to one conversations. But depending on the nature of a meeting, often lots of these happen between pairs of people in the meeting.