Meetings are products

I think most meetings are effectively vibe-coded.

Everyone can create a meeting, so everyone does. Someone thinks that "people need to get aligned," so they drop a 30-minute block on the calendar and figure the group willl find the answer together, live, in real time. Often without an agenda, a pre-read, or a clear decision that needs to be made. The meeting becomes a wandering search for the truth, a problem that twenty minutes of prep would have solved before the room ever assembled. And because ten minutes of talking costs an hour of everyone else's time, the price of that wandering adds up fast.

This isn't a "meetings are broken" post. Discussion and debate can make magic in the same way that pressure makes diamonds. But if you look at most people's calendars, you'll find a lot of meetings that technically produce output, but that nobody actually designed. They're just there. Recurring. Forever.

The problem is we don't think about meetings as something that needs to be designed at all.

Good software starts with understanding the use case. Who's using it? What problem does it solve? What does success look like? Meetings deserve the same questions, but we almost never ask them. We just throw a recurring block on the calendar and hope it works.

A five-person sync to resolve a bug needs a very different meeting than a five-person cross-functional review of an upcoming campaign launch. The first is about problem-solving together in real time. The second is about making sure the work that's been done is good enough to move forward. But we use the same format for both.

Me time vs we time

Part of this is being honest about what kind of work the meeting is actually for. I think about this as the difference between "me time" and "we time." (h/t @ryandawidjan)

Me time is where deep work happens. Reading, processing, forming a perspective. You don't need a room for that. You need space and focus. We time is where you bring what you've processed and work through it with others: sharing perspectives, surfacing disagreements, making decisions together. The reason we send pre-reads is to protect this boundary. The pre-read is me time. The meeting is we time.

When meetings collapse the two, you end up with a room full of people processing live, out loud, for the first time.

Who, what, how, and when

At the beginning of the year, we decided to revisit a lot of our operating cadences, and it's pushed me to think about meetings through four questions: who, what, how, and when.

Who needs to be in the room? This is both about name and role. Do you have people who can both weigh in correctly and make decisions appropriately?

What are you trying to accomplish? If you can't answer this in a sentence, you probably shouldn't send the invite. I like Hiten Shah's framing that meetings should compress uncertainty. What specific uncertainty are you compressing? What should be true after this meeting that isn't true before it?

How should the participants interact? Is this a working session toward a specific outcome or a review of a specific artifact? Are there rules on who can speak? Do you need a facilitator? A note taker? Or is this ephemeral? Are you just playing jazz together?

When should this happen, and how often? Almost every recurring meeting is, at some level, a backstop. It's an else statement that exists because decisions couldn't get made or information couldn't flow well enough through other channels. That's fine (sometimes), but the right frequency depends on how often the work actually needs that fallback. If nothing accumulates between sessions that requires live resolution, the meeting is too frequent. If every session is overflowing, it's not frequent enough or your async channels aren't working. And regardless of cadence, never wait for the meeting. If something can be resolved before you're all in the room, resolve it.

When these four things are aligned, meetings feel like the most valuable use of your time. When they're not, meetings feel like a tax.

Diagnosing the problem

A simple way to pressure-test this is before you send a meeting invite, try completing this sentence:

"This meeting will be a good use of time if [who] can [what] by [how]."

If you can't complete it, that's a signal on what you haven't thought through. Can't name the who? You don't know who actually needs to be there. Can't articulate the what? You don't know why you're having a meeting. Can't describe the how? You're about to vibe-code it and hope for the best.

Vercel previously had a Monday meeting to review what was launching that week. If you filled in the sentence, it would read: "This meeting will be a good use of time if leadership can build confidence in the week's launches by reviewing what's shipping." Sounds right. But the who was off. Leadership was reviewing launches without the people who actually owned them. And because the who was wrong, the how broke down. The meeting would surface questions that nobody in the room could answer, which kicked off a bunch of follow-up Slacks to track down the people who could. A thirty-minute meeting was producing hours of extra work.

The fix wasn't to cancel the meeting. It was to rewrite the sentence. We expanded the who to include the product leads and that naturally changed the how from "review the list and follow up later" to "have the conversation with the people who can actually speak to it, right there." More expensive on paper. But (hopefully) cheaper in total because the conversations actually resolve in the room.

We're still iterating on it. That's what treating a meeting like a product actually means. You ship it, you get feedback, and you keep making it better.

Tend to your meetings

I encourage everyone to tend to your recurring meetings the way you'd tend to any product you're responsible for. I think you'll find meetings that were set up months ago for a team that's since doubled with no structural changes, or meetings that just limp along because nobody's thought to question them.

And look, I know some meetings exist because the highest-paid person in the room decided that's how things work. Organizational gravity is real. You can't always redesign from scratch. But you can usually adjust the who, what, how, or when within the constraints you have.

Meetings are a product. Treat them like one.