How far in advance should I book a flight?

For domestic and short-haul flights, 15 to 21 days before departure is the sweet spot in most conditions. For international long-haul, 40 to 75 days out performs better because those fares are priced off more demand signals. Peak holiday travel needs 60 to 90 days because inventory at those dates sells out earlier regardless of price.

Is Tuesday the best day to book flights?

The evidence for this is weak and the origin is a 2010 study that airline pricing systems have long since replaced. Current fare data shows only a small mid-week advantage, and it does not reliably outweigh the risk that a fare you saw on Tuesday is gone on Wednesday. Book when your dates are right and the fare is good, rather than waiting for a named weekday.

Do fares really drop in January and February?

Generally yes for peak destinations, because those months sit outside the holiday and school-break calendar in both hemispheres. Booking a spring-break trip or a European summer holiday in January is the cheapest window for those specific dates. Booking peak summer dates for a Mediterranean destination in February does not help, because the demand for those dates is already at its maximum.

What is the cheapest way to find a price drop?

Price alerts on Google Flights, Hopper, and Kayak work because they watch the fare rather than only the calendar. What they cannot do is predict how low it will go. Treat an alert as a prompt to look, not as a guarantee. Also set an alert on the route without dates, which reveals the seasonal floor price for that city pair across the whole year.

Does booking a refundable fare ever pay off?

Rarely, and rarely at the price difference quoted. A fully refundable fare typically costs 1.5 to 2.5 times a non-refundable fare on the same flight. The better option is a fare with a 24-hour cancellation right, which US federal rules grant on bookings made directly with the airline at least seven days out. That gives you the flexibility without the premium.

Why do some fares drop after I booked them within 24 hours?

This happens on flights with more seats than demand, and it is the airline reclaiming the difference rather than a pricing error. Airlines usually detect this themselves and void the price. It is not a strategy that works reliably, and deliberately mis-stating your booking details to attempt it risks a chargeback, a cancelled ticket, and being banned from the fare bucket for a year.

When Is the Best Time to Book Flights? (2026 Data)

At a glance
15-21 days out for domestic and short-haul, 40-75 days for long-haul international.
Peak holiday travel: 60-90 days, because inventory closes before price does.
The Tuesday-night booking rule is a myth left over from 2010-era pricing systems.
Book a non-refundable fare on JetBlue, Delta, or Alaska — US rules give you a 24-hour cancellation window, which is most of the benefit of refundable at none of the cost.
Set price alerts on the route without dates to see the seasonal floor for that city pair.

Fare patterns last verified October 2026. Airfares reprice constantly — treat these as a booking window, not a price quote.

The booking window, by route type

The single most useful variable is how long the route is. Short-haul fares are driven by close-in demand and airline seat deployment, so they hold up well and then drop sharply in the last three weeks. Long-haul fares are driven by more inputs — cabin load forecasts, fuel, currency — and behave more like a curve that bottoms out well before departure.

Adding the calendar layer produces the practical answer. Peak holiday dates are priced against total willingness-to-pay, which is high and inelastic, so fares there stop improving long before the generic window would suggest.

Route type	Best booking window	Earlier than that	Later than that	Why
Domestic, 500-1,500 miles	15-21 days	High and flat	Steep climb after day 10	Capacity decisions happen close in
Short-haul international	20-35 days	Moderate	Rises sharply inside 14 days	Border and seasonal demand shifts later
Long-haul international	40-75 days	Roughly flat to slightly high	Rises after day 30	More demand inputs, slower repricing
Peak holiday (Thanksgiving, Christmas, July 4)	60-90 days	High	Sells out, then rises	Inelastic demand closes inventory early
Shoulder season leisure	30-50 days	Good	Flat	Low demand means fewer last-minute sales
Open-jaw and multi-city	45-80 days	High	Rises early	Complex itineraries are priced as a package

Season matters as much as lead time

For any given route, the cheapest month to book a fare is almost never the cheapest month to fly. The gap between them is the discount, and it exists because airlines price by competition — when several carriers fly the same route, fares fall; when one carrier has a monopoly, they do not.

Two patterns hold up well in current data. Midweek departure is cheaper than the equivalent Saturday flight, often by 15-30% on a route with competing carriers. Out-of-season dates price well below the same route in peak, but only on routes where seasonal demand actually moves — a ski destination in July is not a bargain.

Booking period	What is happening	Fare behaviour	Advice
Nov-Dec (for next summer)	Early load factors still soft	High, with wide variance	Track only; do not buy yet
Jan-Feb	Post-holiday demand trough	Seasonal lows for the year	Best window to book any spring or summer date
Mar-Apr	Summer demand firming up	Rising steadily	Book summer trips now or accept the climb
May-Jun	Peak inventory closing	High and firm	Only shoulder dates are still cheap
Jul-Aug	Summer peak	Highest of the year	Wait for the September drop for next year
Sep-Oct	Post-summer, pre-holiday	Sharp dip, then rise into Thanksgiving	Best window for winter and early spring travel

The myths that cost people money

Tuesday at midnight is the cheapest time to book. This traces to a 2010 study measuring fare loads, not fares, and the systems that produced those patterns no longer price this way. Current data shows a small mid-week edge at most. Waiting for a specific night to buy a fare you saw on Monday is a net loss.
Booking on a specific site always finds the cheapest fare. OTAs and airline sites share the same inventory. The variable fare differences come from fare-class buckets and ancillary fee strategy, not from the booking channel. Sometimes the airline is cheaper, sometimes it is not, and no site reliably wins.
A drop right after you buy means you can game this. On a flight with spare seats, the airline will often cancel a sold-out fare bucket. Detecting it yourself and filing a claim is technically possible but results in chargebacks, cancelled tickets, and a permanent fare-bucket ban. Air Canada's 2016 attempt at exactly this produced a backlash and they abandoned it.
Refundable fares protect you. The price difference is 1.5 to 2.5 times the non-refundable fare, which is more than most trip cancellations cost. A 24-hour free cancellation on a US booking is the version of this protection that is actually worth paying for.
Clearance and last-minute deals exist. Airlines rarely discount unsold seats on flights that are close to full, because they would rather sell them at a higher fare in the last three weeks. Genuine last-minute drops happen when a schedule change creates an empty section.

What actually works: alerts, routes, and flexibility

Three practical tactics beat any calendar rule. Set a route-level alert on Google Flights with no dates set, which shows the lowest fare that route hits across the whole year and reveals the seasonal shape. Check the multi-city form even for a round trip, because the two-leg form prices differently and occasionally returns a lower total on the same flights.

The third is flexibility, priced explicitly. A Google Flights date grid or a Hopper matrix shows the fare on every nearby date at once. Being willing to leave a day earlier or fly a different airport is usually worth more than any booking-window rule. In a single search, the flexible-date view typically finds a fare 15-25% below the cheapest-looking option on your original dates.

Tactic	Typical saving	Effort	Works for	Free tool
Book 15-21 days out on a domestic route	10-25% vs booking early	Set a reminder	Flexible travellers with fixed dates	Any price alert
Flexible-date grid	15-25% on the same route	One search	Anyone without a hard date	Google Flights date grid
Nearby airport check	$40-$160 per person	One search	Routes with 2 airports within 90 minutes	Multi-airport option on any metasearch
Route-level annual alert	Reveals the seasonal floor	One alert	Trips you plan months ahead	Google Flights, without dates
Weekday shift	15-30% on competitive routes	High	Flexible travellers, non-peak routes	Date grid
Airline-direct booking for 24h cancellation	Not a saving — risk reduction	None	US bookings 7+ days out	US DOT rule

The short version

Put a reminder in your calendar at 21 days for domestic travel and 55 days for long-haul international, and add 75 days for holiday dates. Ignore the weekday myths. When the reminder fires, check a flexible-date grid and a nearby-airport search before you commit, and book direct with the airline if you want the 24-hour cancellation window.