a field guide
the meeting playbook
running fewer, better meetings across hybrid and distributed teams.
start with: does this need a meeting?
by shanyce lora
this opens in your browser. “save your copy” gives you a working file to keep offline.
contents
what's in here
pages you turn. a last page you reach.
01 · why meetings feel so expensive
the evidence behind meeting fatigue
here's the part i don't want to skip: there's real research behind this, and what to measure on your own team.
i think most people treat an open slot on someone's calendar as an invitation to fill it, whether or not that person is mid-thought. free time isn't a suggestion. none of us practice great calendar hygiene, honestly, that's its own book. but a lot of what's broken about hybrid meetings starts right there, before anyone logs on.
cognitive load and video exhaustion
researchers at stanford's virtual human interaction lab (bailenson, 2021) found four things that make video calls tiring: staring at faces up close for too long, the extra effort of reading nonverbal cues on a screen, not being able to move around, and the mirror effect, where you're constantly watching yourself instead of the room.
calendar debt and context switching
microsoft's work trend index found a "triple peak day," a third burst of productivity late at night as people try to recover the deep work hours meetings ate during the day. researchers at uc irvine (mark, gudith and klocke) found it takes about 23 minutes and 15 seconds to get back to full focus after an interruption.
the cost of a bad meeting
steven rogelberg's research at unc charlotte, done with otter.ai in 2022, found that a company with 5,000 employees wastes more than $100 million a year on meetings that didn't need to happen. atlassian's 2024 survey of 5,000 knowledge workers found meetings miss their stated goal 72 percent of the time.
if one person's remote, everyone's remote
before any of this, i was a nanny. i heard "that's not fair" about a hundred times a day, and the fix was almost always the same: everyone gets the same rules. gitlab's remote-first playbook runs on that same logic. if one person joins a meeting remotely, everyone treats the whole meeting as remote. skip that, and you get a two-tier meeting, where the people in the room get the context and the influence, and everyone else watches.
02 · does it deserve to exist
book a meeting like it's capital spending
meetings cost real hours from real people. every hour you put six people in a room is six hours you can't spend somewhere else. only book one when working async can't get you the same speed, alignment or trust.
is this a status update or announcement?
yes convert to async. post it to the wiki or a loom, track it on the project board.
no, keep going ↓
does this need real debate, live consensus or trust-building?
no async review. circulate the doc, open a 48-hour comment window.
yes, keep going ↓
are the conditions met? clear agenda, essential people only, pre-read sent
no push back or fix it first. yes book the sync.
run this as a tool · three questions, then a verdict.
6 meetings worth keeping
- decision-making (daci). brings a driver, approver, contributors and informed people together to resolve a trade-off that's already on the table.
- retrospectives. keep these, but don't let them turn into a round robin where everyone politely says what went well. the useful part is figuring out what actually broke, and whether the system made that outcome inevitable.
- brainstorming and problem-solving. some problems move faster when the right people can interrupt each other. this is one of the only meetings where i want the mess.
- kickoffs and cross-functional alignment. skip this one and you'll spend the whole project re-litigating who owns what.
- interactive all-hands / exec q&a. the one meeting where it's fine for a leader to say "i don't know yet." that's kind of the point.
- 1-on-1s. the first meeting people cancel when they get busy, which is usually exactly when they need it most.
3 meetings worth killing
- status check-ins. everyone goes around the room giving updates. fix: automate it through your project tracker or a short written summary at the end of the week.
- broadcast and announcement assemblies. one person talks, everyone else listens. fix: record a three-minute video and drop it with a doc for questions.
- zombie recurring meetings. outlived their reason to exist, and only survive because they're on the calendar. fix: cancel them, or bring them back on a shorter cycle if you actually need them.
03 · make hybrid actually hybrid
neutralizing distance bias
i've sat in enough of these to know the real damage happens after the meeting ends. the call drops, and the people still in the room keep talking. that's where the actual decision gets made, and the person who just logged off has no idea it's happening.
one-in, all-in. if anyone joins remote, everyone in the room opens their own laptop and talks through the room mic. everyone stays active in chat, the whiteboard and reactions, in person or not. i'm also anti forced camera, honestly. unless there's a real reason you need to see someone's face, nobody should be pressured into it. demanding it doesn't make a meeting more honest, it just makes it more uncomfortable.
the first-call protocol. call on remote people before the room gets a chance to. it's a small move that fixes a lot.
the remote advocate. give this to an actual person. "be mindful of remote folks" never works on its own, someone has to own it.
- room mic covers everyone clearly, no echo.
- remote audio plays through room speakers at equal volume.
- one shared doc or workspace for everyone.
- no physical whiteboards unless captured on camera.
- everyone in the room has their own laptop for chat and files.
- every video feed is scaled the same.
04 · across time zones
the async handoff
real-time work slows down across time zones. the fix is a clean handoff. zone a writes the rfc during business hours, defines the choices, and posts it by end of day. zone b reads the doc during theirs, adds written input, and finalizes next steps.
gitlab's remote architecture
if the decision only exists in someone's memory, it isn't a decision yet. it doesn't count until it's written down in a public project space.
shape up, from basecamp
if you can't write the pitch as a memo, you don't understand the problem well enough yet to have the meeting.
the 48-hour window
when you ask distributed teammates for input, give them 48 hours before you close out feedback. otherwise the decision favors whoever's in your time zone.
05 · make it count
structure beats rambling
when a sync meeting is the right call, run it with structure. that's what keeps thirty minutes from turning into an hour of opinions.
the silent meeting
i'm not against small talk, but forced platitudes at the top of a call slow things down. it's 2026, nobody needs three minutes of "how was your weekend" to prove they're friendly. swap the slide deck for a written memo instead. everyone reads it together, in silence, commenting in the margins, before anyone talks. record the call, too. i've been burned by too many "i never said that" conversations to skip this step. it's not about catching anyone out, it's just insurance.
| time | what happens | |
|---|---|---|
| 0:00–0:05 | facilitator frames the memo's core objective. | |
| 0:05–0:20 | everyone reads and comments in silence. | |
| 0:20–0:40 | discuss the top unresolved comments. | |
| 0:40–0:45 | assign action items, log next steps. | |
the daci worksheet
daci gives a decision explicit owners: a driver who curates options and drives alignment, one approver who makes the final call, 2 to 5 contributors whose expertise shapes it, and anyone informed once it's done.
your calendar is supposed to support the work.
it shouldn't become the work.
the tools that compute live next door: the decision tree, the calendar debt audit, the 24-hour pre-flight, the scorecard.
this file is yours. "save your copy" hands you the field guide itself, and it keeps working with the wifi off.