Inside Jettova's Planning Room: How a Group Actually Decides Where to Go

If you've ever tried to plan a trip with friends in a group chat, you already know how it goes. Someone tosses out a destination, someone else counters with a different one, three people stop responding, and a week later the trip is dead. The problem isn't the friends — it's the channel. Group chats are bad at decisions. Jettova's planning room is the part of the product designed specifically to fix this. This is a walkthrough of how it works, end to end.

Step one is creating the room. Anyone in the group can start it: hit /room/create, give the room a name (something like 'Bachelorette Lisbon' or 'Adam's birthday weekend'), and the system gives you a short invite code. You share that code or link with your group via whatever chat you already use. There's no account requirement to join — anyone with the link can drop in immediately. The owner is just the person who created it; they don't have any special planning powers later.

Step two is each member casting their vibes and budget. When someone joins, they see a clean form asking three things: what kind of trip do you want (a multi-select of vibes — beach, nightlife, cultural, adventure, wellness, history, party, food, romantic, nature, shopping), what's your budget range per person, and which cities you'd specifically like to visit (optional). Members fill this out individually. There's no chat, no negotiation, no waiting for consensus before answering — just personal preferences in a structured form.

Behind the scenes, the room aggregates these inputs in real time. The vibes get scored: each member's picks count, and the most-voted vibes float to the top of the consensus. The budget overlap shows the floor (lowest member's max — the planning ceiling everyone has to respect) and the median, so the group can see immediately whether they're aligned on a $1,500 trip or a $4,000 trip. The destinations members suggested go into a candidate pool. All of this is visible to every member as soon as the votes land. There's no 'submit' moment — it's continuous.

Step three is the destination shortlist. Once the group has cast vibes and budgets, the room produces a shortlist of six destinations matching the consensus mood (the most-voted vibe bucket: beach-party, adventure, cultural, or wellness). The shortlist is selected with weighted-random sampling — destinations matching more of the group's vibes are favoured, but not deterministically, so two groups picking the same mood see different cities. This matters: without it, every beach-party group would see Aruba first every time, which is both boring and probably wrong for half of them. Members then thumb-up or thumb-down each destination on the shortlist. Live consensus shows which one the group is converging on. Anyone can also suggest a custom destination (a city not on the shortlist) and put it to the same vote.

Step four is the build. When the group settles on a destination, one member (typically the most engaged person, but anyone can do it — there's no admin role) clicks 'Build the itinerary'. This launches the Jettova concierge flow with the destination, mood, and group budget already locked in. The builder can't override any of those — the back button stops at the storyboard step, so the votes the group cast are protected from being unilaterally changed.

The storyboard is the heart of the trip-planning experience. Instead of asking the AI for one big itinerary, the builder picks one day at a time. Each day starts with a vibe choice (recover, explore, cultural, adventure, party, foodie, nightlife, etc.) — and the AI generates four activities and meals in the chosen vibe, each at a real venue with a real price. The builder can swap individual activities, drop in a real bookable Viator activity from the gallery shown alongside, or regenerate the whole day. When the day looks right, they hit 'Love it!' and move on to the next.

Meanwhile, the rest of the group is watching live. Each approved day shows up in their room view in real time. Members can veto any activity by tapping it — the activity gets flagged, and the builder sees the veto count inline. This is where the group dynamic stays healthy. Members aren't passive; they're actively shaping the plan as it gets built. The builder isn't pushing through everyone's preferences alone — they're surfacing flags and addressing them.

When the last day is approved, the builder lands on a final review screen showing the entire itinerary expanded by default. Every vetoed activity is flagged with the veto count and a one-tap Swap button — the builder can replace any flagged activity right there, on the review screen, without backing up. Real bookable Viator activities can be dropped onto the last day from a gallery at the bottom. When the trip looks good, the builder hits Book This Trip.

What happens next is the part most group-trip tools get wrong. The trip is locked in — itinerary, hotels, activities — but each member books their own flight from their own city, on their own dates if they want different ones. When members open the trip page, they see a prominent 'Your booking details' card asking for their departure city and travel dates (defaulting to the group's). Once filled in, a Book My Flight button runs a fresh single-passenger search via Duffel and routes them to the in-app checkout. They enter passenger details and pay with their own card via Stripe. They get their own booking reference. The group's trip is shared; the bills aren't.

The whole flow takes 30 minutes for a small group, maybe an hour for an indecisive group of eight. The version of this trip in a group chat would have taken three weeks and probably died. The structural difference isn't AI; it's that every step has a defined input format (vote, vibe, veto), so the group's preferences actually compound into a decision instead of dispersing into a thread.

If you want to try it, /room/create takes about ten seconds. Most groups find that the first time they use Jettova together, they spend the most time on the destination vote (because nobody has used a tool that just shows live consensus before, so everyone wants to play with it). After that, the storyboard build is fast. The booking is faster. And the trip happens.