about
Group trips die in the gap between intent and starting point.
Longtable is the product we built so they don’t.
The thesis
The household
is the unit.
Individual-centric planning tools fail for multi-household trips because they model the wrong unit. A weekend with three families has six adults and maybe eight kids, but it doesn’t have fourteen users. It has three households, each with its own calendar, its own budget ceiling, and its own division of labor between the person who plans and the person who just wants to show up.
The calendar is the bottleneck. The partner who never opens the doc is the silent constraint. The planning energy gets spent re-deciding things that were already decided, because the decisions are scattered across four threads and nobody knows which one is the source of truth.
Longtable reasons over the intersection of constraints, not each person’s preferences individually. Each household submits once. The feasible weekend locks itself. One page is the source of truth from kickoff to memory card.
what exists today
The tools weren’t
built for this.
Google Flights finds fares for one person. Airbnb shows cabins for one household. Doodle asks when you’re free but doesn’t know what to do with the answer. Splitwise settles bills after the trip. None of them coordinate the whole trip for the whole group.
Otto
A co-planner,
not a chatbot.
Otto is a two-phase research engine. Phase one maps the possibility space — destinations that fit the vibe, dates that clear every household’s calendar, lodging within budget. Phase two synthesizes three concrete options with real places, real pricing, and honest reasoning about the trade-offs between them.
It’s not generating filler. It’s doing the research a good planner would do, in five minutes instead of five hours. The output isn’t a list of links — it’s three ready-to-decide trip options with dates, lodging, anchor activities, and cost per family.
Powered by Claude, Anthropic’s frontier AI models.
Three commitments
What the product refuses to be.
Households, not individuals.
Every planning tool treats each person as a user. A trip with three families isn't shaped that way. Each household submits its constraints once — calendar, budget, kid logistics, the partner who just wants to be told when to show up. The product reasons over the intersection. The Reluctant Participant is a first-class user, not an afterthought.
A co-planner named Otto.
Type a paragraph. Three real trip options come back — destination, dates, lodging in budget, cost per family. The point isn't the AI. The point is that the planning artifact exists before the group decides whether to plan. Most trips die in the gap between intent and starting point. Otto closes that gap in five minutes.
A reason to plan another.
When the trip is over, a memory card. Six months later, that card surfaces with one prompt: plan another? If the friendships matter, the answer is usually yes. The product isn't trying to be daily. It's trying to be the right tool, twice a year, for fifteen years.
where this goes
Consumer trips first.
Corporate retreats, weddings, reunions next.
Longtable starts with the trips that matter most — friend-families, annual reunions, the gatherings that survive the year. The same coordination engine scales to corporate retreats, wedding weekends, and any group that needs to get from intent to itinerary.
Built with Next.js · Supabase · Claude · Vercel
The trips that almost didn’t happen,
happen.
Longtable is in closed alpha. Seats open one household at a time, by invitation.