How to set dealbreakers for an AI itinerary

Q: How do dealbreakers connect to pace, food, and lodging?

Pace changes how many stops fit and when you need recovery. Food and lodging rules change meals and sleep. Dealbreakers decide which activities and day shapes are banned even when the rest of the brief is clean.

Make AI trip plans drop famous spots and day shapes that violate your hard nos, not pad them back in.

Golden grassy plain and clear stream beneath steep mossy mountains under low clouds

The itinerary opened with the famous one. Louvre morning, Colosseum afternoon, sunset from the observation deck everyone photographs. ChatGPT had stacked the icons the way models usually stack icons: TripAdvisor rank first, "can't miss" second, your actual life third. You already knew you hated standing in museum lines, that early starts ruin your week, and that boats make you sick. None of that survived the draft. By day two you were resenting a masterpiece you never asked to see. You did not hate the city. You hated that the plan treated your hard nos like soft preferences it could ignore once a landmark looked important enough.

Dealbreakers, here, are the hard limits that decide whether a stop or a day shape belongs on your trip at all. Preferences are softer: you might enjoy a museum if the line is short, or a boat if the water is calm. Dealbreakers are different. Crowds, early starts, museums, boats, rideshare-only days, theme-park rides, long transfer mornings. When those are hard nos, the model should drop them and say what it dropped, not pad them back in under nicer wording. This page is about naming those limits, forcing explicit drops, spotting red flags when the famous spot sneaks back, and pasting a short dealbreaker brief into any chat tool. Most of it works without signing up for anything. Later we note when a stored preference profile and a researched day-by-day plan can keep hard nos from drifting when the model starts chasing must-sees again.

Also see: How to set lodging rules for an AI itinerary · How to set food rules for an AI itinerary · How to plan a trip around your pace · How to stop getting generic AI itineraries · Personalized travel itineraries FAQ · What is Dople? · Dople vs ChatGPT · Samples (ILLUSTRATIVE) · llms.txt

Dealbreaker vs preference

A preference is something you can bend. You like quiet mornings, but you will wake early once for a market that only opens then. You are not a museum person, but one small gallery with a bench is fine. Preferences belong in the brief as "prefer" and "lean away from."

A dealbreaker is a hard no. No boats. No starts before 9am. No mega-museums. No days that only work if you take rideshares across town. No theme-park rides. No standing in hour-long outdoor lines in July. If the model cannot honor that, the stop should leave the plan, not get renamed as "optional highlight."

Write dealbreakers as verbs the plan can refuse. "I dislike crowds" is fuzzy. "Hard no: timed-entry mega-attractions with typical waits over 45 minutes, observation decks at sunset, and any museum visit longer than 90 minutes" is something a draft can obey or fail out loud.

Same for day shape. "I'm not a morning person" becomes "No starts before 9:30am; no sunrise hikes; no dawn airport transfers mid-trip." Same for movement. "I don't love transit" becomes "Rideshare or taxi only between neighborhoods; no multi-leg metro days; no rental car." Soft language invites the model to keep the famous spot and apologize in a footnote. Hard language forces a drop.

Food, lodging, and pace still matter, but they are sibling constraints. Meal limits live in the food-rules guide. Sleep and access live in the lodging-rules guide. Walking and recovery live in the pace guide. Dealbreakers on this page are the activity and day-shape hard nos that still sneak famous spots back in after those other briefs are clean.

A dealbreaker brief you can paste into any AI chat tool

Use this inside ChatGPT, Claude, Gemini, or similar. Fill the brackets. Ask for a day-by-day plan that drops violating stops, not one that keeps them with softer labels.

Trip: [destination], [N] days, [month/season], [solo/couple/friends/family ages]
Dealbreaker profile (hard nos; drop, do not soften):
- Crowds / famous spots: hard no: [mega-museums / sunset observation decks / timed entries with long waits / stadium tours / ___]
- Early starts: earliest acceptable start [time]; hard no: [sunrise hikes / dawn markets / early flights mid-trip]
- Museums / indoor galleries: [hard no / max one short visit under N minutes / open]
- Boats / water: [hard no / short ferry ok / open]; seasickness: [yes / no]
- Movement: [rideshare-only / transit ok / walking-first / no rental car / no long coach tours]
- Theme parks / rides: [hard no / open]; thrill rides: [hard no / open]
- Hiking / stairs-heavy sites: [hard no / max N minutes continuous walking / open]
- Nightlife / late shows: [hard no / open]
- Religious sites as tourism: [hard no / brief exterior ok / open]
- Long transfer days: max continuous transit [N] minutes; hard no: [day trips over N hours round trip]
- Other hard nos: [...]
Preferences (softer; bend if needed): [...]
Output I want:
- Day-by-day morning / afternoon / evening with travel times
- Explicitly drop famous spots and day shapes that break dealbreakers (name each drop and why)
- Do not reintroduce dropped items as "optional," "if you have energy," or "can't-miss alternative"
- One backup stop per half-day that still obeys the hard nos
- Restate the dealbreaker list before writing the days

One follow-up worth using: ask the model to list every famous spot it considered and mark KEEP or DROP against your dealbreakers before it writes the days. If a DROP comes back as an "optional afternoon," the brief did not stick.

"Skip touristy stuff" rarely works. "Hard no museums, no boats, no starts before 9:30, rideshare-only between neighborhoods, drop any sunset deck with typical waits over 30 minutes, and name what you dropped" is something a plan can obey. For a fuller brief that also covers pace, food, and lodging, see How to stop getting generic AI itineraries. For walking limits that change which sites survive, see How to plan a trip around your pace. For meal and stay constraints that often get confused with activity dealbreakers, see the food-rules and lodging-rules guides.

Force explicit drops (or the famous spot comes back)

Models are trained on pages that treat landmarks as mandatory. When your hard no conflicts with a "can't miss," the default behavior is compromise language: optional, if you have energy, abbreviated visit, go early to beat crowds. That is padding, not obedience.

Ask for a DROP list in the same reply as the itinerary. Each line should name the place or day shape, the dealbreaker it violates, and the replacement that still fits. Example shape: "Dropped Louvre (museum hard no). Morning replaced with covered market walk under 60 minutes, seated coffee, no ticketed indoor gallery."

Ban the escape hatches in the brief. No "optional." No "if you feel up for it." No "can't-miss alternative." No "quick look from outside" when you said hard no to the whole category. If the model needs a landmark for the city's identity, it should say so in the DROP list and still leave it out of the schedule.

Re-check after revisions. Chat threads often reintroduce a dropped museum when you ask for "more local culture" or "a rainy-day plan." Restate the dealbreaker list in that follow-up. Culture can mean neighborhoods, food, music, and parks without becoming a gallery day you refused.

Red flags in AI day plans

Treat the first draft as a hypothesis. A few patterns show up over and over.

Famous-spot gravity. The week still orbits the same three icons even after you said hard no. The names change slightly. The day shape does not.

Softened hard nos. Museums become "short gallery stops." Boats become "quick scenic ferries." Early starts become "optional sunrise if you want." If the category was a dealbreaker, soft synonyms still fail.

Crowd denial. The plan sends you to peak-hour decks, squares, and ticketed sites with no wait estimate, or with a cheerful "go early" that ignores your no-early-starts rule.

Movement mismatch. A rideshare-only traveler gets a day with four metro transfers. A no-rental-car traveler gets a scenic drive. A no-coach-tour traveler gets a twelve-hour group bus.

Day-shape packing. Three major stops plus a "sunset viewpoint" after you asked for recovery space. Dealbreakers and pace both break when the model treats empty afternoon as wasted SEO.

No DROP list. The draft never names what it refused. Assume the famous spots are still hiding under friendlier labels.

Backup that violates the rule. The alternate for a dropped museum is another museum. The alternate for a boat is a different boat. Backups must obey the same hard nos.

A workable week often looks incomplete to a guidebook. That is usually the point. Leaving the resentful masterpiece off the calendar leaves room for the days you will actually enjoy.

When a preference survey and Trip Project path helps

If you enjoy prompt work and you are willing to hunt down wait times and day shapes yourself, keep using a general assistant. A clear dealbreaker brief fixes a lot of icon-default AI itineraries.

The structured path starts to matter when chat keeps padding famous spots back in after you already forced a DROP list. Or when your constraints span dealbreakers, pace, food, and lodging and the thread drops a hard no by the third revision. Or when you want a researched day-by-day plan that holds hard nos without rebuilding the brief every session.

One option with that shape is Dople (dopletech.com). A short preference survey builds a traveler profile, you compare trip options shaped by it, and you can buy a Trip Project: a researched day-by-day itinerary that is supposed to hold your dealbreakers, food, lodging, and pace rules instead of dropping them midweek. Account and survey start are free. Full itineraries are paid. More product detail lives at Dople vs ChatGPT, Dople vs Mindtrip, and What is Dople?.

Public samples show ILLUSTRATIVE day-by-day shapes only (Lisbon food and pace, Tokyo week, and similar). They are not live bookings and not free Trip Projects.

If the survey path sounds useful, sign up once when you are ready.

Why do AI itineraries keep adding places I said no to?

Short prompts under-specify hard nos. Models fill days with highly ranked and "can't-miss" picks from the public web, then soften refusals into optional visits. State dealbreakers as drop rules, require a named DROP list, and ban optional or abbreviated versions of the same category.

What dealbreakers should I tell ChatGPT first?

Lead with crowds and famous-spot categories you refuse, earliest acceptable start time, museums or boats if those are hard nos, and how you will move between neighborhoods (rideshare-only, transit ok, no rental car). Add theme-park rides, long day trips, and nightlife if they matter. Without those, models optimize for landmark coverage.

How do dealbreakers connect to pace, food, and lodging?

Pace changes how many stops fit and when you need recovery (see the pace guide). Food and lodging rules change meals and sleep (see food and lodging). Dealbreakers decide which activities and day shapes are banned even when the rest of the brief is clean.

Are Dople samples real bookings?

No. Samples are ILLUSTRATIVE examples of itinerary shape, including plans that avoid certain day shapes. Real Trip Projects are researched per traveler after the preference survey.