manzil Beta
Notes on Travelling Together

The group chat is where the trip lives, not where it gets planned

The trip group chat is where the energy is: the memes, the countdown, the photo dump. It is also, structurally, a terrible place to keep a plan, because a chat only knows how to remember things in the order they were said, not in the order that actually matters once you're standing at the airport.

This page is published by the developer of manzil, which makes an app for holding a trip's plan. It argues about what a chat thread is structurally good and bad at, not that your group should stop using one.

There’s a lot to love about the trip group chat. It’s where the trip actually starts, weeks before anyone books anything, with someone sending a random reel and someone else replying “we HAVE to go.” It’s where the countdown lives, where the packing panic happens, where the best photo from day two gets sent at 1am with six replies of laughing emoji. None of that should change. A trip without a chat group behind it is a trip with less joy in it, genuinely.

But somewhere around week two of planning a trip with friends, the same chat gets asked to do a second job it was never built for: holding the plan itself. The hotel booking. The driver’s number. Who’s paying for what. The final headcount. And this is where things quietly start going wrong, not because anyone in the group is careless, but because a chat is structurally the wrong shape for this second job.

A chat only knows one axis: time

A group chat remembers things in exactly one order: when they were said. That’s a fine way to hold a conversation and a bad way to hold a plan, because a plan needs to answer “what’s the current answer” and a chat can only answer “what was said, and when.” The booking confirmation someone pasted on a Tuesday is still in there, technically, but it’s now under four hundred messages of trip banter, memes, and someone asking what to pack. Finding it means scrolling, or asking, “wait, did someone book the hotel?” in a chat that already has the answer, buried, twelve days back.

A decision made in chat has no state

This is the quieter problem, and it’s the one that causes actual damage on the trip itself, not just annoyance beforehand. When six people discuss where to eat on night two, and three of them say “sure, works for me,” has the group decided, or just discussed? A chat has no way to mark the difference. There’s no line between “this is what we’re doing” and “this is one option someone floated that got a few thumbs up.” Everyone in the thread is left to infer it, and different people infer differently, which is exactly how a group ends up standing outside a restaurant that half of them thought was settled and half thought was still open.

“Everyone is typing” isn’t consensus, and a string of thumbs-up reactions to a message isn’t a decision either, however much it might feel like one in the moment. A real decision needs a place to land that isn’t just further down the same scroll, something that says, plainly, this is what we’re doing, not this is what we talked about.

The specific things that get lost

Ask anyone who’s run a group trip what actually went missing, and it’s almost always the same short list. The hotel confirmation number, needed at check-in, buried somewhere in a chat with no signal in the hotel lobby. The driver’s phone number, sent once, on day one, now impossible to find fast when the group is standing at a pickup point wondering where the car is, which is half the reason it’s worth agreeing on a fallback before the signal drops. The running list of who’s paying for what, which existed at some point as a message but was never a document anyone could actually check against, is how a group ends up with a friend who never settles up: the debt started as an unspoken assumption instead of a written number. And the poll: someone ran a poll, six people answered it, and then nobody actually counted, because a poll result sitting in a chat isn’t the same as a decision anyone acted on.

None of these are big failures. They’re small, and that’s exactly the problem. A group trip rarely gets derailed by one big thing going wrong. It gets derailed by five small ones in the same afternoon, each of them recoverable on its own.

Chat for conversation, something else for state

The actual fix isn’t to abandon the group chat, and nobody who’s tried that has kept the group happy for long. It’s to split the two jobs the chat was doing. Let the chat keep doing what it’s genuinely good at: the conversation, the excitement, the running commentary that makes a trip feel like a trip before it’s even started. And give the plan itself, the hotel, the driver, the money, the actual decisions, somewhere that isn’t scroll-shaped. Somewhere a person can open and immediately see the current answer, not the history of how the group got there.

This is the specific gap manzil’s trip board is built for: a place for the booking details, the who’s-paying-what, the poll result that actually got read, sitting apart from the conversation but attached to the same trip, so nobody has to scroll to find out what was decided. It’s deliberately not a second chat. No reactions, no threads, nothing to keep up with, just the plan, sitting still. It’s invite-only for now, so you’ll need a code from someone already using it.

The group chat did its job the moment it got everyone excited enough to actually book something. It was never supposed to also be the filing cabinet. A trip goes more smoothly the moment those two jobs live in two different places: the noise stays in the chat, where it belongs and where it’s genuinely good, and the plan lives somewhere that isn’t going to bury the hotel confirmation under four hundred messages about what to pack.