The Best Time to Book Flights in 2026, According to the Data

• 38 days

  Average low point for U.S. domestic fares

• $130

  Average saving from booking domestic 15–30 days out versus 6+ months out (Expedia)

• $190

  Average international saving from booking 31–45 days out versus 6+ months out (Expedia)

• ~3%

  What the cheapest booking day of the week is actually worth

Every year a handful of large travel companies publish a study telling you when to buy your ticket, and every year the numbers do not match. One says 38 days. One says two weeks. One says three months. They are all reporting honestly and they are all measuring something slightly different, which is why this page gives you the windows and an account of where they came from — because knowing that a number is a median of searched fares rather than a median of paid fares changes what you should do with it.

The short version: booking windows are real, they differ by region far more than by season, and they matter roughly ten times more than any day-of-week ritual. Below is every number worth having, followed by the reasons they disagree.

The booking window for every route type

This is the whole answer in one place. Everything after it is the reasoning.

Why a booking window exists at all

Airlines do not price seats. They price inventory buckets, and the booking window is a side effect of how those buckets are managed.

A flight goes on sale roughly eleven months before departure, when the schedule is loaded. At that moment the airline knows nothing about how this particular flight will sell, so it opens conservatively: a small number of seats in the cheapest fare classes, most inventory held in the higher ones. A revenue management system then watches the booking curve — how fast seats are selling relative to the same flight last year, adjusted for events, competitor capacity and a dozen other inputs — and reallocates.

If the flight is selling behind curve, the system releases more cheap inventory. If it is selling ahead, it closes the cheap buckets and the fare steps up. This is why a fare can drop and then rise again, and why two people on the same flight can pay $180 and $640 without either having done anything clever or foolish. The full account of that machinery is in how airlines actually price a seat.

The window exists because the airline needs time to learn. Too early and it has no information, so it holds price. Too late and the remaining buyers are the ones who cannot shop around, so it raises price. The middle is where it is still trying to fill the aircraft and still has enough runway to discount.

The three phases of a fare's life

Every fare passes through the same three phases. Recognizing which one you are in tells you what to do far more reliably than counting days.

The practical upshot is that the anchor phase and the squeeze phase are both expensive, for opposite reasons. Almost all of the value is in the middle, and the middle is short.

U.S. domestic: 21 to 52 days out

Analysis of Google Flights historical fare data puts U.S. domestic prices at their lowest between 21 and 52 days before travel, with the average bottom at 38 days. That is a five-week zone, which is generous enough to plan around without being so vague as to be useless.

Expedia’s 2026 Air Hacks report, which measures tickets actually purchased rather than fares observed, lands tighter and later: it found the most affordable domestic economy bookings at 15 to 30 days out, averaging about $130 less than tickets bought more than six months ahead. Both findings point the same direction — the expensive mistake is booking too early, not too late — while disagreeing about exactly where the floor sits.

What this means in practice

• Start tracking about three months out. You are not shopping yet. You are learning what the route normally costs, which is the only way to identify a discount when one appears.
• Expect the useful zone between about seven and three weeks before departure. Check every few days rather than obsessively; nothing you do changes the price.
• Treat 14 days as a hard line. Inside it, fares generally only go up. If you are still unbooked at that point, buy the least bad option and stop optimizing.

International, region by region

“International” is not a useful category. A flight to Cancún and a flight to Auckland have nothing in common commercially, and their booking windows are months apart. Use the region.

Mexico and the Caribbean

These behave much like domestic routes: short, frequent, heavily contested, and flown by carriers with leisure-heavy demand they need to fill. The low zone runs roughly 26 to 68 days out, centering near 44. Because these are the classic winter-escape markets, the competitive pressure inverts seasonally — the same route that discounts hard in September will not discount at all in February.

Europe

Transatlantic pricing is the most seasonal thing in commercial aviation. Google Flights data points to trips to Europe from the U.S. being cheapest around 94 days before departure on average, with prices generally lowest from roughly 48 days out. Expedia’s booked-fare analysis finds international savings of about $190 for tickets bought 31 to 45 days out versus six months or more.

The wide spread between those numbers is not noise; it is summer versus winter. A July fare to Rome has no reason to discount at 48 days because the aircraft is already full. A November fare to Rome has every reason to. If you are flying transatlantic between May and August, behave as though the window is earlier and shorter than the averages suggest.

Asia and Oceania

Five to seven months out. These routes have fewer daily frequencies, higher aircraft utilization and far less spare capacity to dump, so the discounting behavior that produces a late window elsewhere barely happens. There is also less competition per city pair, which is the underlying reason.

Does the day you book matter?

The most durable myth in travel is that Tuesday at 1pm is when fares drop. It had a basis once: airlines filed fares through a clearing house on a weekly schedule, and Monday evening filings became visible on Tuesday. That schedule stopped governing anything years ago. Repricing now happens continuously.

Expedia’s 2026 analysis found the cheapest day to book is Friday, and the effect is worth about 3% against Sunday, the most expensive day to buy. That is a real, measurable difference, and it is also the smallest number on this page.

The day you fly matters far more

The day you travel is a genuine lever, because it changes which passengers the airline is competing for. Business demand clusters on Monday mornings and Thursday and Friday evenings. Leisure demand clusters on weekends. Tuesday and Wednesday are what is left.

The international finding is the surprising one. Friday used to be an expensive departure day because it was the end of the business week; as corporate travel patterns shifted, Friday long-haul demand softened enough to flip it into the value slot. It is a good reminder that these rules describe current airline behavior rather than laws of nature, and that any guide older than a couple of years is probably telling you something that stopped being true.

Returns follow the same logic. A Wednesday return prices well in both directions, and a Sunday evening return is consistently the most expensive seat on the itinerary.

Months, seasons and the shoulder trick

Seasonality is worth more than the booking window and much more than the booking day, because it changes the underlying supply and demand rather than nudging the allocation.

The full month-by-month picture is its own subject — the cheapest time of year to fly, by destination has the regional calendar and the national figures. The short version is that the weeks immediately either side of a peak season are dramatically cheaper than the peak itself, while the weather and the daylight barely change. Europe in late September and early October is the clearest example: the crowds are gone, the fares have collapsed to winter levels, and the temperature is within a few degrees of August. The Caribbean in early December, Japan in November, and the U.S. in the last two weeks of January follow the same shape.

Holidays, where the window does not apply

For Thanksgiving, Christmas, New Year and spring break, the booking window does not apply, and treating it as though it does is one of the most expensive errors in this subject.

The window exists because airlines discount to fill seats. On dates where demand is known months in advance and exceeds supply, there is nothing to fill. Fares open high and go higher. There is no phase in which the revenue management system is worried about that flight.

• Domestic holiday travel: book two to four months ahead.
• International holiday travel: book four to six months ahead.
• Flexible on dates: departing on the holiday itself — Thanksgiving Day, Christmas Day — is reliably the cheapest day in the period, often by a wide margin.

Why the major studies contradict each other

This is the section most guides leave out, and it is the one that makes the rest usable. The major airfare studies disagree because they measure different populations.

Once you see this, the contradictions resolve. A study saying “book two weeks out” is describing when its price-sensitive customers bought. A study saying “book 38 days out” is describing when the price was lowest. Those are different claims and both can be true at once.

The practical synthesis: use the fare-history windows to decide when to start watching, and use the booked-fare windows as a reminder that waiting slightly longer than feels comfortable is usually not punished — right up until the squeeze phase, where it always is.

Why the perfect window is the wrong goal

Every number on this page is an average across an enormous number of routes, and the variance around those averages is larger than the averages themselves. A specific flight from a specific airport on a specific date may bottom out at 90 days or at 12, and no study can tell you which.

Which means optimizing the timing of a purchase is the wrong goal. The right goal is to be watching a route long enough, and broadly enough, that you see the drop when it happens. Timing is a proxy for attention, and attention is the thing that actually pays.

This is also why searching well matters more than timing well. The traveler who watches four routes for two months will beat the one who watches a single route and buys on exactly the statistically correct day, every time.

How to apply all of this in ten minutes

The whole thing, as a procedure.

1. Identify your region from the table at the top and note the two numbers: when to start watching, and the cheap zone.
2. Look up the route in Google Flights and read the price graph. Write down what normal costs. This single step does more than everything else combined.
3. Set a price alert on the route, from every airport within two hours of home.
4. Ignore it until your watch date. Checking daily changes nothing and makes you more likely to buy out of fatigue.
5. When a fare arrives more than 25% under normal — and here is how to work out what normal is — book it that day, regardless of where you are in the window.
6. If you reach 14 days out unbooked and you must travel, buy. The squeeze does not reverse.
7. Use the 24-hour cancellation rule as a free hold on anything you need a few hours to confirm — and as the way to take cash back if the fare drops after you book.

Booking windows are worth knowing and they are not worth agonizing over. They tell you when to pay attention. What you do with the attention is the part that saves the money.