About Booth

Last updated 19 August 2026

Booth is a free web app for deciding where a group goes out. You start a plan, drop its link in the chat, and everyone suggests, approves, and vetoes real places until the night is decided.

Why Booth exists

Groups are bad at deciding where to go out, and the reason is structural rather than social. A group chat is a single stream with no memory: suggestions scroll away, votes are scattered across forty messages, and anything anyone said last time is gone. Nobody in the thread is doing anything wrong. The tool simply cannot hold a decision, so the decision does not get made until somebody eventually gives up and books something.

Dietary needs make it worse in a specific way. Somebody has to remember who is vegan and who has the nut allergy, then check every candidate against both. That job has no owner, so it lands on whoever cares most, every single time — and when it gets dropped, it gets dropped on the one person who then has nothing to eat. The question that started Booth is the one in the mock-up on the front page, asked in real group chats over and over: wait, is she vegan or vegetarian?

Who it is for

Three to eight friends picking dinner or drinks. Not work teams and not families — the register is social and fast, and the design assumes a phone held one-handed rather than someone sitting down to plan.

It also assumes the group is scattered. Booth was built by someone whose own friend group spans several continents, so somebody weighing in on a Brooklyn dinner may be reading from Berlin at two in the morning. They still need to see the same prices, distances and times as the person standing on the street outside. Participation is bursty: people drop in, vote on three things, and come back after dinner. The room is built to be re-entered, not re-read.

What makes it different

Three things in one room, and it is the combination rather than any one of them:

• Dietary needs are structured, not remembered. Each person sets their tags once, on their own profile, and those tags travel into every plan they join. The group’s combined needs are visible in the room, and every suggestion is checked against them without anyone having to ask.
• The AI is a participant, not a widget. Nomi posts into the same message stream as everyone else, can put places forward under her own name, and can be spoken to directly. She is in the room rather than bolted to the side of it.
• Approval voting, with a veto that costs something. Approve as many places as genuinely work for you. Vetoing requires a written reason and carries your name — so a veto is an argument the group can answer rather than an anonymous number that quietly kills an option.

Put another way: poll apps do voting without knowing anything about food. Restaurant apps know about food but not about your group. Group chats have the group and no structure at all.

What Booth will not do

It does not sell anything, show ads, or take a cut of a booking — Booth does not do bookings. It sets no advertising or tracking cookies. It does not post anywhere on your behalf, and it has no feed, no follower count and no notifications competing for your attention.

Dietary information gets particular care, because it is health-adjacent and people hand it over casually. It is shown to the group you are in and used to check places against, and that is all. The privacy page is written in plain words for exactly this reason, and every claim on it is checked against what the code actually does.

Where it is going

Booth is early and is being built in the open, one piece at a time. The room came first, then Nomi, then reaching her from WhatsApp so you can add a place or approve one without opening anything. What it will not become is a social network for restaurants. The goal is a group deciding in fewer messages than the group chat would have taken, with nobody’s constraint quietly overlooked — and a product that has achieved that has no reason to ask for more of your evening.