When to Book Flights: What the Booking-Window Data Actually Says

Sponsored by FlyraGo, a flight, hotel, eSIM and car rental comparison service. FlyraGo supported the research and had no role in the findings; every source and number below is checked to the publication's own standard.

Every traveller has heard a version of the advice. Book on a Tuesday. Book 47 days out. Never book weekend fares. Wait for a price drop. Most of it is folklore wearing the clothes of data, and the strange part is that real data exists: two of the largest fare databases in the world, Google Flights and Expedia, have studied millions of bookings and published their answers. The answers disagree with each other in interesting ways, and reading why they disagree teaches more about airfare pricing than any single rule could.

What the two big studies found

Google Flights, working from fare data covering January 2021 through August 2025, puts the single cheapest day to book a domestic flight at 39 days before departure, inside a broader favourable window of 23 to 51 days. For international trips it recommends booking 49 or more days ahead, with a 26-to-79-day window for flights to the Caribbean and Mexico.

Expedia's 2026 Air Hacks report, built on 2025 bookings alone, draws the window tighter: 15 to 30 days before departure for domestic economy, a range it prices at roughly $130 below bookings made further out. It also warns about the other edge: travellers who booked more than six months ahead paid about $160 more than those inside its window. For international fares it names 31 to 45 days as the value zone, and finds the boldest savings of all in the last fortnight, 8 to 15 days out averaging $225 below the six month bookers, a bargain for anyone brave enough to cut it that fine.

Two studies, millions of bookings each, and a combined picture that looks like this:

Trip type Google's best window Expedia's best window Where they agree
Domestic economy 23 to 51 days (best day: 39) 15 to 30 days Roughly three to seven weeks ahead
International 49+ days 31 to 45 days Four to seven weeks ahead, longer for complex routes
Caribbean and Mexico 26 to 79 days Not separately published About one to eleven weeks

The overlap is the honest answer. A domestic ticket bought between three and five weeks before departure sits inside both studies' favourable ranges, and neither study's data supports the most common traveller instinct, which is to book the moment flights open for sale.

Why the studies disagree

The disagreement has a cause, and it is instructive. Google's dataset begins in January 2021, when the world was still emerging from the pandemic and travel patterns were reshuffling year by year; fares behaved differently in 2021 and 2022 than they do now, and every booking window derived from that period absorbs some of that distortion. Expedia's report uses 2025 data only, closer to current conditions but a single year, and single years carry their own quirks.

There is also a structural difference worth knowing: Google's analysis runs on the fares it displays in search, while Expedia's runs on bookings completed through its own platform. The two measure adjacent but not identical universes, much as two opinion polls measure the same electorate with different methodologies. Neither is wrong. Each describes its own data, which is why the numbers above stay separate and marked with their sources rather than averaged into a false single answer.

The seasonal exceptions that beat the general rule

Both studies agree that the calendar around the trip matters more than the calendar of the booking. Google's seasonal findings:

The pattern underneath the numbers: the busier the travel period, the more the favourable window shifts earlier, because peak date inventory sells from the top of the price ladder. A Christmas flight bought inside the ordinary 15-to-30-day window is a flight bought after the cheap fares have gone.

The day you fly matters more than the day you book

The most reliable savings in either study come not from booking cleverness but from flying when others do not. Expedia's current report measures Tuesday the cheapest domestic departure day, about 14 percent below Sunday, and Friday the cheapest international departure day. January is the cheapest month for domestic fares; August, counterintuitively, is the cheapest month for international travel, running about 29 percent below December on average.

The booking day folklore, by contrast, has quietly collapsed. For years the Sunday booking rule was real; the current studies price it at 2 to 3 percent, Friday for international and Saturday for domestic, a rounding error on most fares. Anyone rearranging their week to book on a specific day is optimising the smallest variable on the ticket.

A method for your own route, in ten minutes

The studies describe markets; your fare is a specific route on specific dates. The method that converts their findings into a cheaper ticket:

  1. Fix the dates and the airports, then search the full market at once. Aggregators exist precisely so no one has to check airline sites one by one. FlyraGo, which supported the research behind this piece, compares flight options from airlines and travel agencies through its search partner and hands you to the seller with the price you picked, with no markup and no account required; the same comparison logic works on whichever search tool you prefer, and the point is the breadth of the first look.
  2. Check the price today and note it. That number is your baseline, and writing it down defeats the airline industry trick of letting memory do the comparison.
  3. Map your departure onto the windows above. Domestic trip inside a month? You are already inside both studies' range, so the question is whether to buy now or wait. Big holiday trip next season? The seasonal table says your window opened earlier than you think.
  4. Set a fare alert for the route and give it a deadline. An alert without a deadline becomes a traveller watching prices rise past the point where the good windows closed. Ten days out from the edge of the favourable window, buy.
  5. Within 21 days of departure, stop waiting. The last week fare curve rises steeply on busy routes, and the 8 to 15 day savings Expedia documents are a dare, not a strategy.

What the data cannot tell you

Honest limits, because fare data has them. The studies describe averages across millions of routes; individual routes can and do defy them, especially on thin regional schedules where one carrier controls most seats. Fare allocations, the number of seats sold at each price rung, are set per flight and sell out in ways no average captures. Fuel and currency move whole markets between studies' publication and any reader's search: the analysts' standard warning, that fuel prices reprice fares faster than booking windows protect them, applies with full force in any period of oil market turbulence. And no study can see unsettled future demand, the school trip, the cancelled conference, that changes a route's price a month from now.

The findings above are directional, verified against each study's own publication, and worth exactly as much as their fit to your route. Where your route disagrees with the average, your route is the data that matters, which is the reason the first step of the method is a search of the actual market rather than a recitation of the averages.

The takeaway, compressed

Book domestic trips three to five weeks out, international trips four to seven, and earlier still for the winter holidays. Fly Tuesday at home, Friday abroad, and favour January and August over December. Ignore the day of booking folklore, whose edge has shrunk to a couple of percent. Write down the first price you see, set an alert with a deadline, and buy inside the window rather than after it. None of it is magic; all of it is data, and now the data is yours.

This article was produced with the support of FlyraGo, a flight, hotel, eSIM and car rental comparison service, whose support made the research possible. The findings, sources and method are presented to The Meridian Report's editorial standard, and every number above is traceable to the studies listed in the sources.