How to prioritize must-dos in an AI itinerary
Rank non-negotiables before nice-to-haves so AI stops treating every stop as equal and dumping your top sights into leftover scraps.
You named three things you would not leave town without: the viewpoint at golden hour, the market morning, the one museum that made the flight worth it. The draft came back with twelve stops. Your museum landed on the transfer afternoon after bags and a long lunch. The viewpoint got a cloudy mid-morning slot because a cafe crawl wanted the evening light. The market sat on the day it was closed. Everything had equal boldness in the prose. Nothing had rank. You spent the week chasing leftovers and telling yourself the famous list was still "covered."
This page is about priority ranking in AI day plans. Must. Should. Nice. Skip. Not hard vetoes, not buffer minutes, not calendar hours, not ticket slots. Those have their own guides. Here you teach the model which stops own the good day slots and which only fill leftover scraps. Most of it works in ChatGPT, Claude, Gemini, or similar even if you never sign up. Near the end we note when a preference survey that holds ranks across revisions can beat paste-and-hope.
Also see: How to set dealbreakers for an AI itinerary · How to leave open time in an AI itinerary · How to plan an AI itinerary around opening hours · How to fit timed entries into an AI itinerary · How to refine an AI itinerary without starting over · How to plan an AI itinerary when travelers disagree · How to plan a trip around your pace · How to set a walking budget for an AI itinerary · How to protect arrival and recovery days · How to choose bases for a multi-city AI itinerary · How to stop getting generic AI itineraries · What is Dople? · Guides · Samples (ILLUSTRATIVE) · llms.txt
What priority tiers mean (and what they are not)
Priority tiers are ranks for placement. A must owns a good half-day: energy, light, open doors, enough walking budget left. A should matters and can slide a day if a must needs the slot. A nice fills only after musts and shoulds are safe. A skip never enters the draft, even as "optional" or "if you have energy."
They are not dealbreakers. A dealbreaker is a hard veto: no boats, no museums, no starts before 9:30. Priority assumes you still want the stop; it decides whether it gets Tuesday morning or the scraps after a packed transfer. Use the dealbreakers guide for categories you refuse. Use this page for ranking among the places you do want.
They are not open time. Open time is slack between blocks so a late tram does not collapse lunch. Priority decides which blocks deserve the best hours before you even draw the buffers. Pair with the open time guide so musts still keep honest gaps.
They are not opening hours. Opening hours are closed Mondays, last entry, kitchen windows. Priority still needs those facts so a must does not land on a locked door. Pair with the opening hours guide.
They are not timed entries alone. A ticket slot is a clock you bought. Priority decides whether that ticket is worth buying for a must, and whether a nice ever gets a ticket at all. Pair with the timed entries guide.
If you only say "we want museums, food, and views," models often print a flat list where every name looks equally important. If you say "MUST: Viewpoint X at golden hour Day 3 or 4; Market Y morning any open day; Museum Z one full morning. SHOULD: neighborhood walk near lodging. NICE: third cafe crawl. SKIP: observation deck with long waits," the draft has fewer ways to bury what you flew for.
Why AI drafts treat every stop as equal
Completeness bias loves a long list. Twelve named stops look more helpful than five ranked ones with empty afternoon air. Training data rewards covering the postcard set. Rank looks like preference bias until you are the one standing in a scrap slot with your "must" after jet lag and bags.
Famous-spot gravity flattens ranks too. Icons get morning slots by default even when you called them nice. Your quieter must (the market, the specific trail, the one gallery wing) gets leftovers because it is less famous in the training mash. Soft language makes it worse: "highlights include" and "you might also enjoy" never say which highlight owns the day.
Group chats without ranks become negotiation soup. One person lists eight shoulds. Another lists six. The model averages into equal weight and nobody's must is safe. Cap musts per traveler and per day, or the week becomes a pile of competing equals. The travelers disagree guide helps when ranks conflict across people; this page is the ranking language itself.
Regenerate loops often keep the flat list. New lunch name, same buried museum. Without an explicit must/should/nice/skip brief pasted into every refine, the model forgets which stop owned Tuesday morning by revision three.
A pasteable priority brief
Paste this as the ranking source of truth before you ask for morning-afternoon-evening detail. Fill brackets with real musts, shoulds, nices, and skips. Attach open-time, opening-hours, timed-entry, walking, pace, spend, food, lodging, and dealbreaker lines you already trust. Ask for a day-by-day that places musts first into legal, energetic slots, then shoulds, then nices only if space remains.
Trip: [city / region], [N] nights, [month/season], travelers: [N] PRIORITY TIERS (locked; place musts before filling nices): - MUST (non-negotiable; own a good half-day; do not bury in leftovers): - [Stop A]: preferred slot [e.g. golden hour Day 3 or 4 / morning any open day]; reason: [why] - [Stop B]: preferred slot [e.g. one full morning]; reason: [why] - Cap: max [1 / 2] MUST pins per day; arrival / recovery / transfer days: max [0 / 1] light MUST only - SHOULD (important; can slide a day if a MUST needs the slot): - [Stop C]: flexible [morning / afternoon]; drop only if MUST + hours conflict - [Stop D]: ... - NICE (fill only after MUST and SHOULD are safe; first to drop under time pressure): - [Stop E] - [Stop F] - SKIP (do not add, even as optional / if-energy / can't-miss alternative): - [Category or named stop G] - [Category or named stop H] DAY PLACEMENT RULES: - Place all MUST pins first on the calendar, then SHOULD, then NICE into leftover scraps only - MUST pins get legal open hours, last-entry slack, and open-time buffers before any NICE is scheduled - Do not put a MUST on arrival evening, recovery morning, or transfer bag-day unless I mark it light-ok - If two MUSTS conflict on hours or walking, ask which to keep that day; do not silently demote a MUST to leftover - Tag every hard stop with tier: MUST / SHOULD / NICE in the day plan MUST NEVER: - Do not treat every stop as equal weight - Do not bury a MUST after a chain of NICE stops - Do not promote a SKIP as optional, abbreviated, or "if you have energy" - Do not spend a prime golden-hour or open-morning slot on a NICE while a MUST sits in scraps - Do not add new postcard stops that were not on MUST / SHOULD / NICE without asking OPEN-TIME / OPENING-HOURS / TIMED / WALKING / PACE (paste short lines): - Open-time buffers: [paste short open-time lines] - Opening hours / closed days / last entry: [paste short opening-hours lines] - Timed entries allowed: [which MUST / SHOULD pins] - Walking ceiling: [paste from walking-budget brief] - Pace / arrival / recovery / bases: [short lines] OUTPUT I WANT: - Day-by-day with each hard stop tagged MUST / SHOULD / NICE - MUST pins placed first with preferred slots honored or an explicit trade named - Checklist: no buried MUST; no SKIP reintroduced; NICE only in leftover scraps; tiers visible on every hard stop
One follow-up that catches equal-weight failure early: ask the model to list every hard stop with its tier and the day slot it received. If a MUST sits after two NICE pins on a tired afternoon, or a SKIP appears as optional, the brief did not stick. Send that stretch back by name. Demand a priority repair without regenerating the whole trip.
For shared constraint lines, lean on briefs you already trust: dealbreakers, open time, opening hours, timed entries, pace, walking budget, arrival and recovery, multi-city bases, and generic AI itinerary.
How priority sits with open time, hours, tickets, and vetoes
Open time keeps musts reachable. A must at 16:00 with a twelve-minute cross-town fantasy still fails. Rank first, then protect buffers around the must with the open time guide.
Opening hours decide whether a must's preferred day is legal. A market morning must on the closed weekday is not priority; it is a locked shutter. Lock calendar facts with the opening hours guide before you fight over slots.
Timed entries are how some musts get a fixed clock. Buy tickets for musts first. Nice stops rarely deserve a prepaid slot that will guilt you into keeping them when the day runs long. See the timed entries guide.
Dealbreakers cut whole categories. Priority ranks inside what remains. If boats are a hard no, do not list a ferry as a should. If museums are allowed but only one is a must, say so so the model does not promote every gallery. Pair with the dealbreakers guide.
Walking budget and pace cap how many musts fit in a day. Two musts across town may blow the ceiling even if both are "non-negotiable." Raise the ceiling for that day, take transit, or demote one must to should with an explicit trade. Arrival and recovery days usually hold zero or one light must. Multi-city transfer mornings are the same: bags eat the good slot.
When travelers disagree about what is a must, settle ranks before draft one. One shared must list with a cap beats two private lists the model averages into equal weight. Use the travelers disagree guide for the negotiation pattern; keep the tier labels from this page.
Red flags in AI output
Equal-weight prose. Every stop sounds equally "can't miss." No tier tags. If you cannot tell which pin owns the morning, the brief did not stick.
Buried must. Your top sight sits after a cafe crawl, a shop cluster, and a "quick viewpoint" on a tired late afternoon. Promote the must. Demote or drop the nices.
Prime slot wasted on a nice. Golden hour given to a postcard deck you marked nice while your must viewpoint sits at 11:00 in flat light. Swap.
Skip reintroduced. Observation deck, mega-museum, or boat tour you marked skip returns as optional, abbreviated, or "if you have energy." Optional skip is still a skip failure.
Must on a wrecked day. Transfer bag-day, jet-lag arrival evening, or closed weekday. Priority without calendar and recovery rules is still a bad day.
Must pile-up. Three musts on one walking-heavy day with no open-time buffers. Cap musts per day or the ranks become a new denseness problem.
Should quietly promoted. A should grabs the only open morning because it is more famous, while a quieter must gets scraps. Fame is not tier.
Rules live only in chat. By revision three the model forgets which stop was must. Paste the priority brief into every refine until tiers stay visible on each hard stop.
Refine without restarting the whole trip
A buried must rarely needs a full regenerate. Regenerating often reprints the same flat list with a new lunch name and the same leftover museum. Switch to an in-place refine.
Name the buried pin first. "Museum Z is MUST; move it to a full morning on an open day. Drop or demote Cafe Crawl 3 (NICE). Keep lodging and the evening show." Specific beats "make the highlights stronger."
Restate tiers in the same message. Ask for MUST / SHOULD / NICE tags on every hard stop in the revised day. If tags disappear, the refine did not stick.
Trade explicitly when two musts conflict. "Keep Viewpoint X golden hour; move Market Y to the next open morning; do not add a new postcard stop." Silent demotion is how musts become scraps again.
Restate open-time, opening-hours, and timed-entry lines if the bury happened because of a closed door, a zero buffer, or a bad ticket stack. The refine guide has the keep/change/drop pattern. Add one priority line: "Keep locked MUST pins [A, B]; move [C] out of leftover scraps into [preferred slot]; demote or drop NICE [D]; do not reintroduce SKIP [E]; tag every hard stop with tier."
When two travelers argue about whether a stop is must or nice, settle the tier before you edit the clock. The travelers disagree guide helps pick which musts survive the cap.
How this sits with the other briefs
This page is ranking: must, should, nice, skip, and day-placement rules so non-negotiables own good slots. The others keep vetoes, slack, calendars, and tickets from fighting those ranks.
Dealbreakers are hard nos. Priority ranks among the stops you still want.
Open time is slack between blocks. Priority decides which blocks deserve the best hours before buffers are drawn.
Opening hours are closed days, last entry, kitchen clocks. Priority still needs legal doors for must pins.
Timed entries place ticket slots. Buy them for musts first; nices rarely need prepaid guilt.
Pace sets week-level density. Priority caps how many musts fight for the same energetic mornings.
Walking budget caps distance when two musts sit across town.
Arrival and recovery keep first and last days light. Priority usually forbids heavy musts on those days.
Multi-city bases lock where you sleep; transfer mornings are bad must slots.
Travelers disagree settles whose must wins when lists conflict. Generic AI itinerary work is the fuller stack before draft one. Refine is the edit loop after a day still buries a must.
When a preference survey and Trip Project path helps
If you like prompt work and you will keep pasting the priority brief yourselves, a general assistant plus the must/should/nice/skip rules above is enough for many trips.
The structured path starts to matter when chat keeps treating every stop as equal after you locked tiers. Or when musts, open-time gaps, opening hours, and timed tickets live in one messy thread and drift by revision four. Or when you want a researched day-by-day that holds ranks without re-teaching the model every session.
One option with that shape is Dople (dopletech.com). A short preference survey can capture priority-style constraints alongside pace, food, walking, and lodging, you compare trip options shaped by it, and you can buy a Trip Project: a researched day-by-day meant to hold those preferences across revisions so you are not stuck reprinting equal-weight lists. Account and survey start are free. Full itineraries are paid. More detail: Dople vs ChatGPT and What is Dople?.
Public samples show ILLUSTRATIVE day-by-day shapes only. They are not live bookings and not free Trip Projects.
If the survey path sounds useful, sign up once when you are ready.
What are priority tiers in an AI itinerary?
Priority tiers rank what the model places first: must (non-negotiable, good day slots), should (important but movable), nice (fill only after musts), and skip (do not add). Without ranks, drafts treat every stop as equal and bury your top sights in leftover scraps. Dealbreakers are hard vetoes; priority is ranking among stops you still want (see dealbreakers).
How is priority different from dealbreakers, open time, or opening hours?
Dealbreakers drop categories. Open time is buffer minutes between blocks. Opening hours are whether the place is open that day and until what clock. Priority decides which wanted stops own the good half-days before slack and calendars are applied. You usually need ranks plus those sibling briefs.
What belongs in a must-do priority brief?
Must list with preferred day or half-day slots, should list that can slide, nice list that drops first, skip list that never enters, caps on musts per day, and must-never lines that forbid burying musts after nices. Paste open time, opening hours, and timed entries rules alongside it.
Should I regenerate when the draft buries a must-do?
No. Regenerating often reprints the same equal-weight list with a new lunch name. Refine in place: promote the must, demote or drop a nice, restate day-placement rules, and keep locked pins (see the refine guide).
Are Dople samples real bookings?
No. Samples are ILLUSTRATIVE examples of itinerary shape. Real Trip Projects are researched per traveler after the preference survey. More context: Personalized travel itineraries FAQ.