Start here
The shape of the trip
Before a single trick, the boring win almost nobody takes: the same cities in a different order, on slightly different days. It is usually the biggest number on the page.
01
Visiting the same cities in a different sequence, staying a different number of nights in each, changes what you pay in airfare — often by hundreds of dollars, for exactly the same trip.
No riskRuns on every planUsually the biggest winNo trade-off
How the planner finds it
- Every order of your cities is enumerated. Six cities is 720 orders, seven is 5,040. A must-start or must-end city cuts the list before anything is priced.
- For each order, a small dynamic program walks the window day by day and picks the cheapest stay length in every city, inside the minimum and maximum nights you set.
- Prices come from a matrix that already holds every city pair on every day of your window, built from two free fare grids before the optimizer runs — so “one more night here for a cheaper flight there” is decided on real fares, not a per-route average.
- Ties are nudged toward using your whole window instead of collapsing every city to its minimum. The nudge is deliberately tiny — under a dollar across the whole trip, however long the window — so any real fare difference wins outright.
- The three cheapest orders are kept. The recommended one is priced leg by leg with a live search; the runners-up get their long-haul legs checked so the comparison is honest.
Worked example
The example plan: six cities, three orders kept
from the example plan- Order chosen
- Dallas → London → Paris → Budapest → Athens → Rome → Madrid → Dallas
- Flights, per person
- $1,214
- Runner-up order
- $1,215, a dollar more
- Third order kept
- $1,236, $22 more
- Nights it picked
- 3 in Athens, 2 in each of the other five, inside a 2–3 range
On this trip the top two orders came out a dollar apart, and the plan prints that rather than dressing it up: here the order barely mattered and you can pick on taste. The gap is what you are being told, and it grows with the number of cities and the slack in your window.
When it does not work
- Two cities and fixed dates leaves almost nothing to optimise. The win grows with the number of cities and the slack in your window.
- Locking the order, or pinning a start and end city, means you are asking for a specific trip rather than the cheapest one. Legitimate — the plan just costs more.
- Nights are allocated from departure dates at day granularity, so the night spent on an overnight transatlantic leg is charged to the city you have not landed in yet. The optimizer solves in those terms, because that is what your own minimums are expressed in — but the calendar file, the printed sheet and the itinerary card all count from the flights instead, and say so where the two differ.
- More than seven cities is not supported: the number of orders grows faster than any budget for pricing them.
Risk
No risk
One booking, normal rules. Nothing here can be taken away from you.
On your plan: “Your route” — the first section of your plan: the recommended order with every ticket open, and the runner-up orders underneath with the price gap to each.multi-city routingtrip orderingcheapest route between cities
02
Moving one flight by a couple of days — letting one city get longer while the next gets shorter — can drop that leg's fare without changing the trip you are taking.
No riskFree fare gridsOne-tap re-planNights spelled out
How the planner finds it
- Before the optimizer runs, every city pair on your route is priced on every day of your window from two free grids: Kiwi's date-range search (one call per nine-day chunk) and Ryanair's own cheapest-fare-per-day calendar for the routes it flies.
- Once the plan is built, each flight leg is re-read out of that grid three days either side of the date it landed on.
- A move is only reported when it saves at least $12 and at least 10% of that leg's fare, and only the single best move per leg is shown.
- The consequence is stated in nights, not dates: “Madrid 4 nights instead of 3”. The city before the moved flight gets longer, the city after gets shorter.
- Moves that push a city under one night, collide with the leg before or after, or land you home after your return date are dropped. A move that leaves the nights range you set is still shown, labelled as leaving it.
- One tap re-plans the whole trip with those nights, reusing the fares already found, so it takes seconds rather than another full search.
Worked example
One leg, three days later
example output- Flight
- Madrid → London, moved from the 8th to the 11th
- Fare
- $126 → $86
- Saving
- −$40 per person
- Cost in nights
- Madrid 4 instead of 3, London 2 instead of 3
The optimizer already picked the cheapest dates inside the nights you allowed. What is left in this section is the trade you never authorised it to make — so it asks instead of taking it.
When it does not work
- Fixed dates with tight nights ranges leave nowhere to move. The section will be empty and will say so.
- The grids only cover pairs Kiwi and Ryanair price. A leg neither covers gets no cheaper-days check at all.
- Savings under $12, or under 10% of the fare, are not shown. Chasing $6 across six legs is not worth a re-plan.
- Grid fares are cached up to 12 hours and are not a booking. The re-plan re-checks the legs that matter.
Risk
No risk
One booking, normal rules. Nothing here can be taken away from you.
On your plan: “Cheaper days”, directly under your route, with a one-tap re-plan on every row.flexible datesfare calendarcheapest day to fly
More trip, same money
Free extra cities
Cities you were never going to pay for — hours you already spend in a connection, and whole regions that are cheap only because of where you already are.
03
A one-stop ticket whose connection is long enough to go into town, for the same money as — or less than — the fastest way to fly that leg. One ticket, one bag policy, one airline responsible for you.
No riskOne ticketProtected connectionCounts from your minimum
How the planner finds it
- Every flight leg is searched twice: the normal search, plus a second search with the airline layover filter set to your minimum (10 hours by default; anywhere from 4 to 24 in Advanced). The two result sets are merged and deduplicated by flight number and price.
- Each itinerary's connections are read out. A connection counts only if it meets your minimum and the city is not your origin, your destination or your home city.
- A layover in your own country is not a free city unless you specifically asked for that city.
- Every candidate is priced two ways: against the cheapest fare on that leg, and against buying it as two separate flights (A to the layover city, then the layover city to B the next day).
- It is kept when it costs no more than $60, or 15% of the leg fare, above the cheapest option — whichever is larger. A city you actually asked for is kept even above that, as long as it beats booking two tickets.
- Consecutive legs get a pair check: is there one A → C ticket with a long connection in B that beats your separate A → B and B → C legs? When there is, it replaces both.
- Each result is measured against a table of minimum useful hours per airport — Istanbul 8, Heathrow 6, Charles de Gaulle 7, Beauvais 9, Singapore 4 — so the plan can tell you whether the stop is long enough to be worth leaving for.
Worked example
Dallas → London, twenty-five hours in Madrid
from the example plan- Ticket
- DFW → LHR via Madrid on Iberia, one booking
- Fare
- $533 per person
- Versus the cheapest way to fly that leg
- $43 more
- Versus buying Dallas → Madrid → London as two tickets
- $199 less
- Time in Madrid
- 25 hours: an evening, a night and a full day
Madrid was already on the list, so one ticket can stand in for a separate leg later in the trip. The same run found a second one: 29.7 hours in Budapest on the Athens → Rome ticket, $20 above the cheapest fare on that leg.
The longer version
Airline stopover programs — the multi-day version
No riskA dozen airlines will let you break the journey at their hub for days rather than hours, with no change in airfare, and some will pay for the hotel. Every carrier and hub on your plan is matched against a table of these programs, and a note only appears when both halves line up: that airline, at that hub, on a ticket you are actually being shown.
What the program table contains
from our data- TAP Air Portugal, Lisbon or Porto
- Up to 10 nights, no extra airfare
- Icelandair, Reykjavík
- Up to 7 nights on transatlantic itineraries
- Turkish Airlines, Istanbul
- Free hotel night(s) on qualifying connections; free city tours on 6–24h layovers
- Iberia, Madrid
- Up to 6 nights, no fare change
- Also in the table
- Emirates, Qatar, Singapore, Finnair, Ethiopian, Copa, Air Canada, Etihad
This is the difference between nine hours in a city and four days in it, on the same ticket — the one hack here where the airline is on your side.
The catch. They have to be booked with the ticket, usually direct with the airline; buying through a third party often forfeits them, and terms change constantly, so every note links to the airline's own page.
When it does not work
- Short connections are not free cities, and how short is short depends on the airport: five hours is enough at Schiphol and not at Istanbul, which wants eight. The card compares your gap against that airport's own figure, and against the careful default when it is an airport nobody here has assessed — it will not invent a number for one.
- An overnight connection is a city you cannot use unless you pay for a bed or stay up. It is flagged “overnight” instead of counted as a free day, and what the gap is worth is worked out from when it falls: only the daylight hours count, and only what is left after the airport takes about ninety minutes at each end.
- You have to be allowed into the country. A US passport needs a UK ETA to leave Heathrow and a Türkiye e-visa to leave Istanbul; the plan attaches the official link for every layover country on your route.
- A long connection usually means a longer, more tiring journey, and one delay can eat the whole visit.
- Immigration, a bag re-check on some itineraries and the ride into town take real hours. Budget the airport's minimum, not the raw layover length.
Risk
No risk
This is one ticket. If the first flight is late the airline rebooks you, your bags are checked through, and the only thing you can lose is the visit.
On your plan: “Free cities”, split into cities on your list and bonus cities you never asked for.Setting: Advanced → minimum layover hours (default 10).long layoverstopover cityfree stopover
04
Somewhere that is absurdly cheap from the last city on your route, where three more days plus a flight home from there costs little or nothing more than the return you were already buying.
Low riskMarginal cost, not fareNegative numbers happenBest few verified live
How the planner finds it
- A curated list of about forty destinations — Tokyo, Bangkok, Istanbul, Reykjavík, Marrakech, Mexico City, São Paulo and the like — is priced from your last city and from home, out of cached fares for the right month.
- The “surprise” number is how many times cheaper it is from your route than from home. Under 1.4× is dropped, and anything actually cheaper from home is dropped outright.
- The number that matters is the marginal cost: fare out to the add-on, plus fare home from the add-on, minus the return flight you already had. That is what the extra days really cost.
- A negative marginal cost means the longer trip is cheaper than the shorter one. It happens, and when it does the card shows a minus sign, not a plus.
- Anything over $700 marginal is dropped as an add-on nobody takes, though it can still appear as a “cheap from here” surprise.
- The best few are verified with two live searches each — out and home — and marked verified. The rest stay labelled as cached-fare estimates.
- Add-ons are measured from the last city only, because that is the only place the marginal cost is honest. Other cities on the route can show a surprise price with no marginal number attached.
Worked example
Three days in Lisbon, added to the end
from the example plan- Lisbon from Madrid, where the trip already ends
- $35 per person
- The same city from Dallas
- $591 per person, about 17× more
- Out + home from Lisbon, minus the return you already had
- +$180 per person
- What that buys
- Three extra days and a different country
The $35 fare is the easy half. The number that matters is the $180, because the way home is what most flight sites never price: they show you the cheap hop out and stop there.
When it does not work
- The candidate list is hand-picked. Somewhere genuinely cheap that is not on it will not be found.
- The marginal cost assumes three extra days and a flight home on that date. Change either and the number changes.
- Rows that were not verified live are cached-fare estimates and say so on the card. Check them before you count on them.
- Cheap flights are not a cheap trip: three days somewhere is still three days of hotels, food, and a visa if one is needed.
- Anything already on your route, or in your own country, is excluded — that is not a discovery.
Risk
Low risk
These are ordinary tickets. The only exposure is committing to a separate return before the outbound is confirmed.
On your plan: “Add a city on the way home”, inside “Make it yours”, and again under “Ways home” so a group can split at the end — some fly home, some go on.wildcard destinationscheap add-onextend the trip
Same flights, cheaper
Cheaper ways to buy the same flights
Identical seats on identical planes, bought differently: split in two, started somewhere else, sold as one ticket instead of two, or quoted in another country's currency.
05
Instead of one ticket from A to B, two separate tickets: A to a hub, then the hub to B a day later. Often cheaper, sometimes not, and worth it mainly for the time in the hub — which is only a day there if the gap falls in daylight. Nobody is responsible for the connection but you.
Medium riskNot protectedRisk badge per comboKiwi combos labelled separately
How the planner finds it
- For every flight leg, hub candidates are ranked by detour — how much further A → hub → B is than A → B. A hub only qualifies if the detour is under 1,500 km or 60% of the direct distance, and cities already on your route are skipped, because a day somewhere you are staying anyway is not a bonus.
- Phase one prices those candidates from cached fares, no times needed, and keeps the ones that are not wildly above the through fare. At most one candidate per leg goes through in the first pass, so several legs get a look instead of one leg taking every slot.
- Phase two prices the survivors properly. Kiwi is searched first because it is free, across three times as many candidates; only the best few then get a paid Google search for real times.
- The gap between the two tickets is measured from the real arrival to the real departure. Under three hours inside Europe, or four hours anywhere else, is rejected outright — as is anything over 36 hours. When the two tickets are at different airports in the same city, three hours are added to that floor, because collecting a bag, crossing the city and checking in again as a new passenger is not something three hours buys.
- What the gap is worth is worked out from when it falls, not from whether it crosses midnight: the airport takes about ninety minutes at each end, and only the hours between 09:00 and 20:00 count as time in the city. A three-hour gap that happens to cross midnight is described as a night in the terminal, because that is what it is.
- Risk comes out of that gap: an airport change is high, an overnight gap is medium above eight hours and high below it, otherwise low above four hours (five internationally) and medium under that.
- A combo is kept even when it costs up to $40 more than the through fare, because a day in Budapest for $10 is a good trade and it should be your call, not ours.
- Kiwi's own multi-airline combos are a different animal and are labelled separately: sold as one booking with the Kiwi.com Guarantee, so a missed connection is rebooked by Kiwi. Those have to beat a through ticket by $40 per person before the planner recommends one.
Worked example
Milan → Budapest → Athens on two tickets
example output- Two tickets
- $70 per person
- Nonstop through fare
- $60 per person
- Difference
- +$10 per person
- What the $10 buys
- 22 hours in Budapest
- Risk
- Medium — overnight gap, not protected
Ten dollars for a day in a city is the point, not the saving. This is the shape to look for on your own plan: not every trip has one, and the example plan we publish came back with none, which is what the empty section is for.
When it does not work
- Separate tickets are not protected. If the first flight is late, cancelled or diverted, the second airline owes you nothing and you buy a new ticket.
- Bags are not checked through. You collect them, clear customs, and check in again.
- An airport change inside the hub — Gatwick to Heathrow, Orly to Charles de Gaulle — turns a gap into a commute, and is marked high risk for exactly that reason.
- You must be admitted to the hub country to do this at all, and some airlines want onward proof at check-in.
- Only a handful of candidates are priced live per plan, and phase one leans on cached estimates. A hub with no cached fares may never get a live look.
- Same-day connections are only tried when the first flight lands by 14:00. Otherwise the second ticket is the next day, which is the pattern that gives you the free day anyway.
- Kiwi's guarantee is Kiwi's, not an airline's. It covers rebooking under its own terms; read them before treating it like a normal connection.
Risk
Medium risk
Every combo carries its own low / medium / high badge, computed from the actual gap. Leave a real cushion — a missed second ticket costs more than the combo ever saved.
On your plan: “Two tickets through a hub”, inside “Other ways to buy the same trip”, with both tickets drawn out and the gap in the middle.self-transferself-connectvirtual interlining
06
A cheap domestic hop to a gateway city, then a separate long-haul ticket from there — because the transatlantic fare out of New York or Lisbon can sit far below the one out of your home airport.
High riskFirst and last leg onlyOnly shown if it savesTwo tickets
How the planner finds it
- This only runs on the first leg (home to your first city) and the last leg (last city to home), and only when the trip crosses regions.
- Gateways come from a per-region list: New York, Boston, Washington, Miami, Chicago, Philadelphia, Toronto, Montreal, Los Angeles and San Francisco in North America; London, Dublin, Reykjavík, Lisbon, Paris, Amsterdam, Copenhagen, Oslo, Stockholm, Madrid, Barcelona, Milan and Rome in Europe; the equivalents in Asia, the Middle East, Oceania, South America and Africa.
- Each gateway then runs through exactly the same two-ticket machinery as a hub combo: estimate first, price the best candidates live, measure the real gap, assign a real risk level.
- Same-day connections are allowed here — you are catching a long-haul, not spending a day — with a minimum three-hour buffer.
- Unlike a hub combo, a positioning split is only shown when it actually saves money. Paying more to fly a longer way is not a hack, it is a mistake.
- The low-cost transatlantic carriers — Norse, PLAY, French Bee, Level — appear by themselves in the live search from the gateway. There is no separate list of them to maintain.
Worked example
Dallas → Milan, tested through New York
example output- Dallas → New York
- $260 per person
- New York → Milan
- $448 per person
- Two tickets, total
- $708 per person
- Direct Dallas → Milan
- $514 per person
- Shown on the plan?
- No — the direct fare won
This is the honest half of the feature, drawn to show you what losing looks like: gateways priced, the split beaten by the direct fare, nothing shown. An empty section on your plan means the check ran, not that it was skipped.
When it does not work
- Two tickets, so every self-transfer caveat applies: no protection, no through-checked bags, your problem if the hop is late.
- Gateway airports are usually not your home airport. Getting there costs money and hours the fare comparison does not show.
- A same-day positioning hop is tight for an international check-in by nature. It gets the same four-hour floor an ordinary cross-region combo gets, and no less — it used to get three, which was the wrong way round for the combo whose second ticket is the expensive one. An overnight in the gateway is safer still and costs a hotel night.
- It only ever fires on the first and last leg of a trip that crosses regions. Nothing in the middle of Europe is a positioning flight.
- Only one or two candidates get priced live per plan on a standard search budget.
Risk
High risk
Missing a transatlantic flight because the domestic hop was late is the most expensive thing on this page. Build in a night, or do not do it.
On your plan: “Positioning flights”, beside the two-ticket combos in “Other ways to buy the same trip”, in the same risk vocabulary.Setting: Advanced → Hacks → Positioning flights (on by default).positioning flightgateway cityfly to the hub first
07
Three ways the same flights get re-priced by asking the airline a different question: buy both ends as one ticket, buy a round trip you only half use, or stop asking for every seat at once.
Low riskThree checksShown even when they tieOne rule-breaker, flagged
How the planner finds it
- All three run on the recommended plan, on live searches, for about three to five paid calls in total.
- Open-jaw: when your first and last legs are both long-haul, they are re-priced as one multi-city booking (home → first city, then last city → home) and compared with the two one-way fares in your plan. It is a win at $15 per person; below that the card still appears and says there is no saving on this route.
- Round-trip-cheaper-than-the-one-way: the same long-haul legs are re-priced as round trips with a plausible return date. If a round trip comes back at least $15 under the one-way you need, the card appears with its rules attached — because using it means throwing the return away, which airlines forbid.
- Group size: for groups of four or more, up to three of the pricier legs are re-priced for a single passenger. If one seat prices at least $10 below what your whole group was quoted, the cheapest fare bucket is smaller than your group, and the card tells you to book in two or three smaller groups, smallest first.
- Every card says whether it actually helped on this trip. Ties and misses are printed rather than hidden, because a tie is information too.
Worked example
All three, run on one six-city trip
example output- Open-jaw ticket
- $1,067 per person
- The same two legs as one-ways
- $1,067 per person
- Nine seats together versus one seat
- $514 either way
- Total worth taking
- $0, all three tied
Drawn as the tie, because a tie is the usual answer and the plan says so on three separate cards rather than going quiet. When it is not a tie, the open-jaw undercuts two one-ways outright and the group check is the most valuable line on the page.
One at a time
Open-jaw: into one city, home from another, on one ticket
No riskOne booking instead of two. Legacy carriers often price this well under two separate one-ways, and one booking protects both flights — if the outbound is cancelled, the airline owes you the return too.
The catch. One booking means one set of change rules; two one-ways can be changed or abandoned independently. It needs both ends long-haul, and some low-cost carriers do not sell multi-city tickets at all.
Round trip cheaper than the one-way (throwaway return)
Against airline rulesLegacy airlines routinely price a one-way close to — and sometimes above — the round trip. Buying the round trip and never taking the return is cheaper, and against the fare rules you agreed to. It is shown with its rules, and it is never used to lower your plan's headline total.
The catch. Never in reverse: skipping the outbound cancels every later segment, including the flight you actually wanted. Nothing else may be on the same booking, and doing it repeatedly on one airline can cost you miles or the right to book.
Group size: book in smaller groups
Low riskAirlines sell seats in price buckets. Ask for six and everybody pays the price of the sixth. Ask for two and then four, and the first two may be cheaper. The check runs for groups of four or more on the pricier legs.
The catch. Separate bookings are separate contracts: a cancellation is rebooked per booking, groups can be split, and sitting together usually costs extra.
When it does not work
- The throwaway return breaks airline rules. It sits in the same section as the two legal tricks on today's results page, which is a wart we are fixing — treat it as a rule-breaker regardless of where it appears.
- Open-jaw needs both ends long-haul; an intra-European start and finish never triggers it.
- The group check only runs at four or more travellers, and only on the more expensive legs.
- Every one of these costs a live search, so they run once per plan on the recommended order, not on the runners-up.
Risk
Low risk
Two of the three are the airline's own products. The throwaway return is not, and carries its own red badge wherever it appears.
- Use the outbound only and simply don't show up for the return; nothing else may be on the same booking.
- Airlines' rules forbid it; done occasionally it goes unnoticed, done often on one airline it can cost you miles or future bookings.
- Never do it in reverse: skipping the outbound cancels the rest of the ticket.
On your plan: “Ticketing tricks”, one card per trick, each saying whether it helped on your trip.open jawmulti-city ticketthrowaway ticketing
08
The same seat on the same plane can carry a different price depending on which country's version of the site you buy it from, and in which currency.
Low riskECB rates3% / $10 thresholdFX fee accounted for
How the planner finds it
- Long-haul legs only, and at most two of them, because that is where the gap is big enough to survive a card fee.
- Each is re-quoted as if bought from the foreign end's country, in that country's currency, through the same live search with a point-of-sale override.
- The result is converted to dollars with European Central Bank reference rates fetched fresh and cached for a day.
- It is only reported when the foreign quote is at least 3% and at least $10 cheaper.
- The card does the arithmetic you would otherwise forget, naming what a 3% foreign-transaction fee would eat out of the saving.
Worked example
Dallas → Milan, re-quoted from Italy in euros
example output- US point of sale
- $514 per person
- Italian point of sale
- €442 ≈ $514
- Difference after conversion
- $0
- Shown on the plan?
- No
Drawn as the zero, because zero is the common result: both long-haul legs come back identical once converted and nothing is shown. This is the check that catches the routes where the gap is 15%, and that tells you when it is nothing rather than implying it is not.
When it does not work
- Card foreign-transaction fees and dynamic currency conversion at checkout can erase a small saving entirely. Pay in the local currency, with a card that has no FX fee.
- Some airlines require a billing address in the point-of-sale country. Online travel agents usually do not.
- Exchange rates move between quoting and paying.
- Only long-haul legs are checked, at most two per plan, because each check costs a live search.
Risk
Low risk
Buying from another country's site is ordinary commerce, not a trick against the airline. The exposure is FX fees and a possible billing-address wall.
On your plan: “Ticketing tricks”, with the foreign quote, the conversion and the FX bite all shown.Setting: Advanced → Hacks → Point-of-sale check (on by default).point of sale trickbuy flights in local currencyVPN flight booking
Not everything flies
Beyond flights
On short hops the plane is often the slow, expensive option once you count getting to the airport. And a night spent moving is a hotel night you never book.
09
On the pairs where the ground beats the plane — city centre to city centre, no airport, no bag drop — the planner treats a train, bus or ferry as just another way to fly the leg.
Low risk~60 city pairsAlways badged est.Competes in the optimizer
How the planner finds it
- A hand-built table covers about sixty city pairs where surface transport is genuinely competitive: Eurostar and the TGV network, Frecciarossa and Italo, AVE and Iryo, ICE and Railjet, FlixBus and ALSA overnight coaches, and the Mediterranean, Adriatic and Baltic ferry routes.
- Each entry carries a mode, an operator, a duration and a US-dollar fare band for one adult booked a few weeks out.
- Ground options enter the optimizer as itineraries, so a train competes with a flight on identical terms and can simply win the leg.
- The midpoint of the fare band is what the optimizer uses. The band itself, the operator name and a Rome2Rio link go on the card so you can see live prices.
- Ground fares are always badged “train · est.”, “bus · est.” or “ferry · est.” and never “verified”. They are a band from our table, not a live quote.
- Ferries are a separate switch, because an eight-hour crossing is a different kind of day out from a two-hour train.
Worked example
London → Paris, chosen as a coach
from the example plan- Overnight coach, on the example plan
- $45 per person, 8h 20m
- Eurostar on the same pair
- $60–$180 band, 2h 20m
- Why the coach won
- It also removes a hotel night
- Badge on the leg
- bus · est. + overnight
The optimizer picked the bus over every flight and over the train, on price plus the saved night. The card still shows the Eurostar, because two hours against eight is your call, not ours.
When it does not work
- The table is finite. A city pair that is not in it is only ever priced as a flight, even when a great train exists.
- Fare bands approximate a booking a few weeks out. Rail fares in the UK, France and Spain move enormously with how early you book.
- Ground legs are never “verified”. Follow the link and see the real price before you count on it.
- Ferries get cancelled by weather and rarely have a same-day alternative.
- The time cost is real. A coach that saves $60 and eats a day has not necessarily saved you anything.
Risk
Low risk
The worst case costs you time, not the trip.
On your plan: Inline on any leg it wins, with a train, bus or ferry badge and a Rome2Rio link for live prices.Setting: Advanced → allow ground legs (on), and Hacks → Ferries (on).train instead of flightEurostarFlixBus
10
A red-eye flight, a night train or an overnight ferry is a hotel night you never book. The planner prices that night and lets the leg spend it.
Low riskDefault $50 a nightFeeds the optimizerSwitch it off to disable
How the planner finds it
- A leg counts as overnight when it departs in the evening, arrives on the next calendar day, and lasts at least six hours — after 17:30 and landing by 13:00 for flights, after 18:00 for trains, buses and ferries.
- Every overnight leg is counted, and each is worth one night at your nightly rate: $50 per person by default, editable up to $500.
- The optimizer subtracts that from the leg's effective price, so an overnight option wins whenever it costs at most a night's worth more than the alternative.
- The plan then states the total plainly: how many nights you sleep in transit and roughly what that was worth.
- Turn the switch off and the same legs are still found and labelled — they just stop getting a price advantage.
Worked example
What the nightly rate did to the example plan
from the example plan- Overnight coach, London → Paris
- $45 per person
- Hotel night avoided, at the default rate
- −$50 per person
- Effective cost to the optimizer
- −$5 per person
- Overnight legs on the whole plan
- 2, worth about $100 per person
A leg that costs less than a night in a bed is free travel, which is why the plan sometimes prefers a slower one. Raise the nightly rate if you stay somewhere expensive and more overnights win; set it to zero and none do.
When it does not work
- A night in a coach seat is not a night in a bed. If you cannot sleep sitting up, the $50 is not real money to you.
- You arrive early, tired, and usually before check-in — so you may pay for the previous night anyway just to have a room at 8am.
- Overnight flights count as departing on their date, at day granularity, which can shift a night on paper against how it feels.
- An overnight gap between two separate tickets is not a saved hotel night. It is a gap you have to survive, and it is badged as risk, not as a saving.
Risk
Low risk
The only thing you are gambling is a night's sleep.
On your plan: An “overnight” badge on the leg, plus a line in the summary counting the nights and what they were worth.Setting: Advanced → Hacks → Overnight = free lodging, with the nightly rate beside it (default $50).red-eyenight trainovernight ferry
11
When a trip has several train legs, a Eurail or Interrail Global Pass can beat buying the tickets one at a time — or not, and the plan tells you which.
Low risk2+ train legsReservations includedYouth price shown
How the planner finds it
- The check only runs when your plan has two or more train legs with both ends inside the Eurail network.
- It counts the distinct travel days and the span between the first and last train, then picks the cheapest pass covering both, from the eight Global Pass options — flexi and continuous, adult or under-28 youth.
- Seat reservations are added on top, because on the trains people actually take they are compulsory: roughly $33–$45 on Eurostar, $11–$38 on TGV and Lyria, $11–$14 on Frecciarossa, $11–$16 on AVE, Iryo and Ouigo, up to $6 on Railjet, ICE and EuroCity.
- Italo is priced separately and called out, because the pass does not cover it at all while Trenitalia on the same route does. A timetable row naming both operators — which is how the Italian high-speed routes are named, since both run them — is read as the Trenitalia train a pass holder takes, not written off as uncovered.
- The pass total is compared with the sum of your actual train fares. It is a win at $20 per person; when it is not, the card says to book early point-to-point fares instead.
- Eurail or Interrail is decided by your passport: Interrail for European residents, Eurail for everyone else.
Worked example
The pass table, and what has to be added to it
from our data- 5 days in 1 month, adult
- $345
- 5 days in 1 month, youth under 28
- $259
- 7 days in 1 month, adult
- $410
- Reservations still payable
- $0–$45 per train, per person
- Compared against
- The summed fares of your actual train legs
A pass rarely wins on two or three trains. It wins when the trip is train-shaped — many legs, flexible days — and the youth price is usually what tips it.
When it does not work
- Pass prices in our table are approximate and converted to dollars. Check eurail.com for today's price, and for youth eligibility, which is measured on the first travel day.
- Reservations are not optional on high-speed and international trains, and in summer they sell out well before the train does.
- Buses, Italo and most ferries are not covered at all. If your plan uses them, those fares stay on top.
- You have to activate the pass and log every journey in the Rail Planner app; an unlogged journey is an unticketed one.
- Advance point-to-point fares in France, Spain and Italy often beat a pass day, if you can commit to exact trains.
Risk
Low risk
The worst case costs you time, not the trip.
On your plan: “Ticketing tricks”, whenever the plan contains two or more train legs.Setting: Advanced → Hacks → Rail pass check (on by default).EurailInterrailGlobal Pass
What a headline fare hides
The fine print that changes the price
A $19 fare into an airport ninety minutes from town, with a bag that costs $55, is not a $19 fare. These are the costs the planner adds back before it calls anything cheap.
12
The cheapest fare on a low-cost carrier, and the cheapest “light” or “basic” fare on a legacy one, does not include a checked bag. Across six legs that is another ticket's worth of money.
No riskPer ticket, not per segmentOnline pricesEstimates, not quotes
How the planner finds it
- Every leg's operating carriers are matched against a fee table: Ryanair $35–$60, Wizz $40–$75, easyJet $30–$55, Vueling $30–$50, Norwegian $35–$55, Spirit and Frontier $40–$65, Icelandair, Norse, PLAY and French Bee $45–$70, and European legacy carriers' light fares at roughly $25–$55.
- Bands are per bag, per person, one direction, booked online in advance — the cheapest way to buy one, and still the honest number.
- One fee per ticket, not per segment, because a bag is checked through on a single booking. A leg with two segments on one ticket is charged once.
- Southwest sits in the table at zero, because two bags are still free.
- The per-leg bands are summed into a single trip-level range printed beside the headline fare, so you never see a total that quietly assumes you travel with a backpack.
Worked example
Bags on the example plan: six cities, seven legs
from the example plan- Flights, per person
- $1,214
- One checked bag, every leg
- +$130 to +$230 per person
- As a share of the fare
- 11% to 19%
- If you travel carry-on only
- $0
This is the number that decides whether a $19 Ryanair leg or a $70 legacy leg is actually cheaper. On some itineraries the bag costs more than the flight.
When it does not work
- These are estimates from a static table, not quotes. Airlines re-price bags by route, season and how close to departure you buy.
- Bags bought at the airport routinely cost two to three times the online price. The table uses the online price.
- Fees are per direction; a return with a bag is two fees.
- Carriers outside the table contribute nothing to the total, so the real figure can be higher than the range shown.
- Cabin-bag rules vary just as much and are not modelled at all. On some carriers even the overhead bag is paid.
Risk
No risk
One booking, normal rules. Nothing here can be taken away from you.
On your plan: A per-leg note, plus a trip-level range next to the headline price.baggage feeshand luggage onlybasic fare no bag
13
A cheap fare into an airport ninety minutes outside the city is not a cheap fare. The planner adds the transfer at both ends of every flight before it decides which flight is cheapest.
No risk~80 airportsFeeds the optimizerMoney and hours
How the planner finds it
- A table of about eighty airports carries the realistic way into town: mode, journey time and a dollar band. Beauvais is a 75–90 minute coach at $18–$22, Luton is the DART plus Thameslink at $15–$26, Heathrow is the Elizabeth line at $8–$17, Bergamo is a 50-minute bus at $10–$13.
- Both ends of every flight leg get one; connections are excluded, because you do not leave the airport on a connection.
- The midpoint of the two bands is added to that itinerary's effective price inside the optimizer. That is what makes an out-of-town airport compete fairly, instead of winning on a headline it does not deserve.
- The trip total is reported as both money and time: a dollar range plus the hours you will spend in transfers.
- Airports missing from the table are listed by name rather than silently counted as free.
Worked example
Airport transfers on the example plan
from the example plan- Transfers, both ends, every flight leg
- +$79 to +$229 per person
- Time spent getting to and from airports
- 8h 20m across the trip
- Airports with no transfer data
- None on this plan
- As a share of the $1,214 fare
- 7% to 19%
Up to $229 and a working day, on a trip whose entire airfare is $1,214. That is the argument for letting the optimizer see it: an airport ninety minutes from town is not the cheap option it looks like on a fare grid.
When it does not work
- Public-transport bands assume public transport. Four people with luggage at 6am are taking a taxi, and the taxi is not in the band.
- About eighty airports are covered. Anything else adds nothing and is listed as unknown rather than assumed free.
- Transfer times are typical, not worst case. Engineering works, strikes and overnight gaps in service are not modelled.
- It is a per-person cost that does not scale like one — a taxi split four ways behaves very differently from four train tickets.
- The hours matter as much as the dollars, and only you can price your own time.
Risk
No risk
One booking, normal rules. Nothing here can be taken away from you.
On your plan: Per leg on each flight card, plus a trip-level money and time total.Setting: Advanced → Hacks → Door-to-door (on by default).airport transfer costBeauvaisStansted
14
If you are under about 26, a dozen airlines sell you a different, cheaper fare — and no comparison site will show it to you unless you ask for it by age.
No riskOff by defaultInformationalPer-leg pointers
How the planner finds it
- The carriers on your plan are matched against a list of airlines that publish youth or student fares: the Lufthansa Group (Lufthansa, SWISS, Austrian, Brussels), Air France and KLM, SAS, Finnair, TAP, Aegean, Iberia, Air Canada, Eurowings, Turkish, Emirates, Qatar and Etihad.
- Each match names the age band — usually 12 to 25, sometimes to 29 or 31 for student fares — and links to the airline's own booking page, because the fare only appears once you enter the age in their search form.
- Alongside those, the plan points at StudentUniverse and the ISIC card, which unlock youth fares and rail discounts the airlines do not publish themselves.
- It is informational by design: these fares cannot be searched programmatically, so the plan tells you exactly which legs to go and re-check by hand.
- It is the one hack that is off by default, because it is noise if nobody on the trip is under 26.
Worked example
Which legs to re-check by hand
from our data- Lufthansa, SWISS, Austrian, Brussels
- Youth fares, ages 12–25
- Air France
- Youth fares, ages 12–24, many long-haul routes
- SAS
- Youth fares, under 26, most routes
- Emirates
- Student fares plus extra baggage, ages 16–31
- Everywhere else
- StudentUniverse and an ISIC card
The plan will not price these for you. It tells you which of your legs are on carriers that sell them, so you spend five minutes on the two that matter instead of thirty on all six.
When it does not work
- These fares are not searchable by the sources this planner uses, so nothing here is verified. Every row is a pointer.
- Eligibility moves constantly and varies by route, and some fares want proof of enrolment, not just an age.
- Youth fares are frequently more restrictive on changes and bags than the public fare at the same price.
- They are not always cheaper. On a competitive route the public sale fare often wins.
- Off by default: turn it on in Advanced or nothing appears in this section.
Risk
No risk
One booking, normal rules. Nothing here can be taken away from you.
On your plan: “Before you book”, inside “Take it with you”, as a per-leg pointer with the airline's own link.Setting: Advanced → Hacks → Under 26 (off by default).under 26 flightsstudent flightsISIC card
Handle with care
Against the rules
One trick that works and that every major airline forbids. We show it because you will read about it anyway, we print exactly what you are risking, and we never count it in your total.
15
A ticket to a city beyond the one you want, connecting in the one you want, that costs less than the direct ticket. You get off at the connection and abandon the last flight. It breaks every major airline's contract of carriage, and we show it for interest only.
Against airline rulesNever in your totalCarry-on onlyOne-way only
How the planner finds it
- For each leg landing at a hub, a table of destinations that hub's home carrier serves beyond it is checked — Austin, San Antonio, Nashville and Denver beyond Dallas; Manchester, Edinburgh and Dublin beyond London; Naples, Catania and Palermo beyond Rome.
- Candidates are ranked for free first, on cached fares whose routing shows the ticket really does connect at your city. Anything cached data says is not cheaper is dropped before a single paid search.
- Up to four candidates then get a live search. A result only counts if the itinerary genuinely stops at your city as an intermediate point and beats the normal fare by at least $25 per person.
- What is shown is the whole thing: what you would be ticketed for, where you get off, and exactly which flights you would be abandoning, by flight number.
- It is never added to your total, never used to choose your route, and never counted in “found” savings. Switch it off in Advanced and the section disappears entirely.
Worked example
The two the example plan turned up
from the example plan- Ticketed Dallas → Manchester, get off at Heathrow
- $370 against the $490 London fare, saves $120
- Ticketed Madrid → Austin, get off at Dallas
- $339 against the $403 home fare, saves $64
- What the plan actually books
- The $490 and the $403
- Effect on the $1,214 total
- $0. It is never counted either way
Both of these are real and both are against the airline's rules, so they are printed with the rules first and the price second, and the plan total is the honest one that ignores them. If you take one, you take it knowingly.
When it does not work
- It only works one-way. The moment you no-show, every later segment on that booking is cancelled — including your flight home.
- Carry-on only, always. A checked bag goes to the ticketed destination and you do not.
- You cannot rely on it. Airlines reroute for weather and delays, and a reroute through a different hub means you never reach the city you actually wanted.
- Internationally you must be admitted to the country of the connection, and the airline may ask for onward proof to a destination you have no intention of reaching.
- It gets worse as it gets popular: fares to secondary cities get repriced, and a route that works today may not next month.
Risk
Against airline rules
Airlines have cancelled remaining segments, voided frequent-flyer accounts, billed the fare difference and banned repeat offenders. It is not illegal — it is a breach of contract, and the airline holds all the leverage. We do not recommend it.
- Book one-way only: the airline cancels every later segment on the same booking once you no-show.
- Carry-on only. Checked bags go to the ticketed destination.
- Do not add a frequent-flyer number, and don't do this often on one airline; it breaks their contract of carriage and they can cancel miles or refuse future travel.
- If the airline reroutes you (weather, cancellations), you may be sent through a different hub and miss your real destination.
- International: you must be allowed to enter the country of the connection; the airline may ask for onward proof for the ticketed destination.
On your plan: “Hidden-city fares · against airline rules”, last inside “Other ways to buy the same trip”, never in the total.Setting: Advanced → include hidden-city fares (on by default; switch it off and the section disappears).skiplagginghidden city ticketingthrowaway ticketing
Nothing here is a promise
The empty sections are the point.
On the example plan we publish, four whole sections came back empty and said so on their own cards: no two-ticket combo beat a through fare, no positioning split beat the direct fare, no ticketing trick or rail pass was worth taking, and no cheaper day beat the dates the optimizer had already picked. Every price you do see is stamped verified, cached or estimate, with the day it was seen.
Give it your cities→