← Back to the calculator

Async vs. Meeting: When the Meeting Isn’t Worth It

Most meetings are a message wearing a calendar invite. Here is how to tell the difference.

Here is the question to ask before you book a meeting: could this be a message? If the answer is yes, it should be a message. That single filter, applied honestly, would cancel a large share of the meetings on most teams’ calendars — not because meetings are bad, but because the meeting has become the default container for everything, including things that were never meant to be discussed live. A status update does not need eight people in a room. A question with a written answer does not need a thirty-minute slot. This page is a decision framework for sorting the two: what belongs async, what still deserves a meeting, and how to go async without anything stalling.

What belongs async

Async means the work happens without everyone being present at the same time: a written update, a recorded video, a comment thread. These belong async:

• Status updates. “Here is where my work stands” is information, not a discussion. A written update or a shared board does the job, and people can read it when it suits them.
• Announcements and FYIs. If the message is one-way, the meeting is a very expensive delivery mechanism. Send it in writing.
• Feedback rounds. Reviewing a document, a design, or a plan is usually better in writing: reviewers can think, quote the exact line, and reply at their own pace instead of reacting on the spot.
• Decisions with a written proposal. When the options and the recommendation are written down, people can comment directly on the proposal — which produces a better record than a meeting nobody remembers.
• Anything spanning many time zones. A meeting that is 9 a.m. for you is midnight for someone else. Async is not a compromise here; it is the only format that respects everyone’s day.

The test for this list is simple: if the meeting’s only output is that people “heard” something, it should have been async. Hearing is not a meeting outcome.

What still deserves a meeting

Async is not a ban on meetings. Some things genuinely need people in the same room at the same time:

• Resolving a real disagreement. When two people or teams disagree and the written thread is going in circles, a live conversation can resolve in twenty minutes what a comment thread cannot resolve in a week.
• Brainstorming that needs fast back-and-forth. Ideas that build on each other in real time — where the third idea only exists because of the second — benefit from live energy. Note the qualifier: if the brainstorm could be a shared document with a deadline, use the document.
• Sensitive news. Reorganizations, layoffs, performance problems, bad results. People deserve to hear hard news from a person, and to ask questions immediately.
• Building trust with a new team. Early in a relationship, face time matters. A new team, a new manager, a new client — these are worth a meeting, because trust is built faster in person than in a thread.
• A hard decision with a deadline and conflicting views. When the decision is genuinely contested and the clock is real, a live meeting with a clear decision-maker is the fastest way to close it.

Notice what these have in common: they all involve something that writing cannot do — reading the room, resolving tension, building trust, or deciding under pressure. If the meeting does not involve any of those, it is a candidate for async.

The three-question test

Before you send the invite, ask three questions. If the answer to all three is no, make it async:

1. Does it need real-time discussion? Will someone need to ask a follow-up question the moment they hear the answer? If the answer is a fact or a status, no.
2. Does it need a decision right now? Is there a deadline measured in hours, not days? If the decision can wait until tomorrow, the meeting can wait too — or be replaced by a comment thread with a reply-by time.
3. Would writing it down lose something? Tone, trust, body language, the ability to read hesitation — if none of those matter for this topic, writing it down loses nothing.

Be honest with the first question, because it is the one people lie to themselves about. “We might need to discuss it” is not the same as “it needs discussion.” If nobody has a question yet, there is nothing to discuss — and a meeting booked for a discussion that might happen is a meeting booked for nothing.

How to go async well

Async fails when it is just a meeting that was cancelled with nothing replacing it. It works when you replace the meeting with something better:

• Write the brief first. What happened, why it matters, what you need from the reader. Three short paragraphs beat a thirty-minute meeting that wanders to the same conclusion.
• Record a five-minute video when tone matters. A short screen recording carries nuance a wall of text does not, and the team can watch it at double speed. Record the update instead of booking thirty minutes to deliver it.
• Use a board for status. A shared board your team actually updates replaces the weekly round-robin where everyone reports what they did last week.
• Set a reply-by time. Async stalls without a deadline. “Comments by Thursday at noon” is a deadline; “let me know what you think” is a wish.
• Keep a written record. The update lives in writing, so nobody needs a second meeting to catch up on the first one — and new joiners can read the history instead of being briefed in a call.

The one-week experiment

Do not try to convert your whole calendar at once. Pick one recurring meeting — the weekly sync is the usual candidate — and cancel it for one week. Replace it with a written update and a reply-by time, and ask the team one question at the end of the week: what did you miss? If the answer is “nothing,” the meeting is gone for good. If the answer names something specific, you have learned exactly what the meeting was for, and you can rebuild a shorter one around that one thing. Price the meeting you cancelled first, in the calculator, so everyone can see the saving in dollars — a $800 weekly meeting is $38,400 a year at 48 working weeks, and that number makes the experiment feel like what it is: a decision about money, not about convenience.