How to Find Cheap Flights: The 2026 Playbook

• $428

  Average U.S. domestic round trip, Q1 2026 — the highest first quarter on record (BTS)

• 21–52

  Days before departure when domestic fares are typically lowest (Google Flights data)

• 14%

  How much less a Tuesday domestic departure averaged versus Sunday in 2026 (Expedia)

• 88%

  Share of paired searches where incognito mode showed exactly the same price

Airfare feels arbitrary because you only ever see one number at a time. You look up a route, a price appears, and you have no way to tell whether $612 is a bargain, a robbery, or exactly what that flight has cost every day for six months. That missing context is the entire problem. It is not that cheap flights are hard to find. It is that cheap flights are hard to recognize.

The average U.S. domestic round trip reached $428 in the first quarter of 2026, the highest first quarter the Bureau of Transportation Statistics has recorded. That number gets quoted as bad news, but it is really a statement about the middle of the distribution. Averages rise because the people paying full price keep paying full price. The distance between the average fare and the cheap tail of the same route is wider than it has ever been, and closing that distance is a learnable skill.

This guide is that skill, in order. Seven steps that work, the tactics that do not work and why people still believe in them, the riskier plays and their real cost, and a worked example that ties it together.

The four things that make a flight cheap

A published fare is the output of a revenue management system deciding how many seats in each fare class to release on a given flight on a given day. It reprices constantly, in response to how fast that specific flight is selling relative to how fast the airline expected it to sell. Nothing about it is personal, and nothing about it responds to you. The mechanics of how airlines price a seat covers that machinery properly: filed fares, booking classes, and why a price rises without anybody raising it.

What it does respond to is competition and slack. A route flown by three carriers with spare capacity in a shoulder month is cheap. The same distance flown by one carrier into a slot-constrained airport in July is not. Which means the four things you control are worth wildly different amounts:

Read that table twice. The variable that generates the most advice — which day of the week to click buy — is worth about three percent. The variable almost nobody adjusts, which destination you fly to, is worth up to three hundred. Every step below is an attempt to move your effort from the bottom row to the top.

Step 1: Pick the fare first, the trip second

Most people book a flight in this order: decide on Lisbon, decide on the second week of October, look up the price, wince, book it anyway. At the moment you fixed the city and the dates, you gave away every source of discount you had. What is left is whatever that route happens to cost that week, and you have no leverage over it at all.

The alternative is to reverse the order. Start with a rough shape — somewhere in Europe, sometime in autumn, about ten days — and let the prices choose the specifics. This is the single biggest behavioral change in cheap travel, and it is free.

How to actually do it

Open Google Flights, enter your home airport, and leave the destination blank. Switch to the Explore map. Set the date field to a flexible window — a whole month, or a duration like “one week” within the next six months — rather than fixed dates. The map fills with prices instead of places.

You will notice immediately that geography and price have almost nothing to do with each other. Lisbon and Reykjavík are routinely cheaper from the U.S. East Coast than Denver is. Mexico City is often cheaper than most domestic destinations. Bogotá and Medellín undercut Caribbean beach routes by half. That is competition and capacity talking, not distance.

Step 2: Widen the map around your home airport

A fare exists on a route, not in a city. If you only ever search one airport, you are watching a tiny fraction of the routes that could realistically get you where you are going. Adding a second departure airport within a two-hour drive is the highest-return change most travelers can make, because it roughly doubles the number of routes that can go on sale for you — and sales do not coordinate across airports. The full second-airport method covers the metro-by-metro pairings and the break-even on the drive.

If you live somewhere genuinely isolated — Bentonville, Missoula, Burlington — the same logic applies with a bigger radius. The nearest large hub within a four or five hour drive will show fares your local airport never will, and on a $600 saving a drive stops looking unreasonable.

Step 3: Search the way the tools want to be searched

Search tools are not interchangeable, and the differences matter more than the interfaces suggest. A short, opinionated version of what each one is for:

Three habits that pay for themselves

Each of these is a Google Flights feature most people never open. The full walkthrough of the five tools covers them properly; the short version is here.

1. Use the date grid, not the date picker. Google Flights will show a seven-by-seven matrix of departure and return combinations with prices in each cell. Shifting a departure by one day is frequently worth more than every other tactic on this page combined, and the grid shows you that in one glance.
2. Read the price graph before you decide anything. It tells you what the route has cost over the past several months, which is the only way to know whether today’s number is good. A fare is not cheap because it is low. It is cheap because it is low for that route.
3. Always price the same itinerary on the airline’s own site before booking through anyone else. A $14 saving at a third-party reseller costs you the ability to fix anything when a flight cancels, because the airline will tell you to call the agency and the agency will not answer.

One specific note worth knowing: Southwest fares began appearing on Google Flights in 2024 and on Kayak, Skyscanner and the Expedia family shortly after, ending years of the airline being invisible to comparison shoppers. Southwest still does not always communicate its free-checked-bag policy in those listings, so a Southwest fare that looks equal to a competitor’s is often meaningfully cheaper once bags are counted.

Step 4: Buy inside the window for your region

There is no magic date to buy. There are zones in which discounting is likely, and they differ by region because capacity and competition differ by region. The short version:

The window is a zone to be paying attention in, not a date to buy on. A genuinely good fare inside your window should be booked the day you see it. A mediocre fare should not be booked just because the calendar says you are 38 days out. The full breakdown of booking windows goes through the regional data properly, including why the major studies contradict each other.

Step 5: Stop searching, start watching

Here is the structural problem with searching. A fare is a moment, not a price. Airlines reprice thousands of routes continuously, discounted inventory is often a handful of seats, and the good ones are gone in hours. To catch one by searching, you would have to search the right route on the right dates during the right window on the right day. Nobody does that, which is why almost everyone pays roughly the average fare.

Watching solves what searching cannot. Instead of asking about a price, you register interest in a route and get told when its price becomes unusual. This is the same shift as the difference between checking a stock quote and setting a limit order.

What a useful alert contains

• The comparison, not just the number. $398 means nothing. “$398, normally $890” means everything.
• The exact date ranges that hold the fare. Most discounted fares live in narrow travel windows. An alert without them hands you a puzzle instead of a deal.
• Your airports, not everyone’s. A fare from an airport you cannot reach is noise, and noise trains you to ignore the alerts that matter.
• Speed. A deal delivered a day late is a story about a deal.

Free tracking in Google Flights covers routes you have already chosen, and paid services exist for the routes you have not. Whether one is worth paying for comes down to a single question about your own flexibility, and the answer is no for more people than the industry likes to admit.

Free tiers exist and are worth using. Paid tiers are worth it if you can be flexible about where you go, and are not worth it if you must travel to one city on fixed dates — no service can discount a route nobody is discounting. That is the honest test, and it is the one worth applying before you pay anyone, us included.

Step 6: Compare the real price, not the headline

The cheapest fare on the screen is frequently not the cheapest way to take the trip. Unbundling has made the headline number an increasingly unreliable guide, and the gap shows up in three places.

Basic economy, priced honestly

A $78 headline gap collapses to $30, and for that $30 you keep the ability to change your mind. Basic economy is genuinely the right answer when you are traveling with a backpack and do not care where you sit. It is the wrong answer roughly as often as it is the right one, and the fare display will never tell you which case you are in. The full break-even math, airline by airline goes through the 2026 fee increases and the loyalty changes that moved the threshold.

The other two leaks

• Getting to the cheap airport. A $60 saving out of a secondary field is not a saving if parking is $22 a day for nine days. Count the ground cost before you count the win.
• Self-transfer itineraries. Some search engines stitch together two unconnected tickets and present them as one journey. If the first leg is delayed, nobody is responsible for the second. Look for the words “self-transfer” or “separate tickets” and price the risk accordingly.

Step 7: Treat points as a second currency

Points are a second currency with its own exchange rate, and the reason they matter is that award pricing does not track cash pricing. When a cash fare spikes for a holiday week, the award price on some programs does not move at all. That mismatch is the whole opportunity.

The most useful currencies are the transferable ones — Chase Ultimate Rewards, American Express Membership Rewards, Capital One miles, Citi ThankYou, Bilt — because they can be moved to whichever airline program happens to have a good price on your route. Points locked into one airline can only ever be as good as that airline’s worst day.

The only formula you need

Divide the cash price of the ticket, minus the taxes and fees you would pay on the award, by the number of points required. That gives you cents per point.

• Under 1.2 cents: pay cash and keep the points.
• 1.2 to 1.8 cents: a fair trade. Use points if you have more points than money.
• Over 2 cents: book it with points, and this is usually a premium cabin or a peak-season economy seat.

One warning that saves people real money: do not skip a genuinely cheap cash fare in order to use points. A $312 round trip to Lisbon is a better outcome than spending 60,000 miles on the same seat. Points are for the trips cash cannot buy cheaply, not for the ones it can. The full points-versus-cash calculation works through the baselines and the two costs people forget to subtract.

The tactics that do not work

Airfare folklore is unusually persistent, partly because prices move often enough that any ritual appears to work sometimes. Here is what the evidence says about the popular ones.

The riskier plays, and what they actually cost

These work. They also carry costs that are usually described vaguely, so here they are precisely.

Hidden city ticketing

You want to fly to Dallas. A ticket from New York to Las Vegas connecting through Dallas is cheaper than New York to Dallas, so you book it and walk out of the airport at the connection. This is legal for individual travelers, and a 2024 federal jury verdict in American Airlines’ suit against Skiplagged — $9.4 million, split between copyright damages and disgorged revenue — turned on the site’s use of airline branding, not on whether passengers may do this. Both sides appealed to the Fifth Circuit.

The costs are contractual rather than legal, and they are real. You cannot check a bag, because it goes to the ticketed destination. You cannot do it on the outbound of a round trip, because skipping a segment cancels the rest of the itinerary. Carriers reserve the right to cancel remaining segments and charge the difference to the direct fare, and some programs explicitly reserve the right to close frequent flyer accounts and confiscate miles. Doing it once on a one-way with hand luggage is a small risk. Doing it routinely on an airline where you hold status is not. The full account of what the courts ruled and what airlines can do works through the cases and the expected value.

Mistake fares

Occasionally an airline files a fare it never meant to publish and a $2,000 ticket sells for $280. These are the deepest discounts in travel, and they come with a specific set of rules about whether the airline has to honor them — it does not, under a 2015 Department of Transportation policy — and what you are owed if it cancels. The full explainer on mistake fares covers the booking protocol, including the one mistake that costs people the most money.

Throwaway returns

On some international routes a round trip is cheaper than a one-way, so travelers buy the round trip and discard the return. It works, it breaches most conditions of carriage, and if you do it on a carrier whose program you care about, it is the kind of thing that gets noticed. Treat it as a one-time tool, not a habit — and check first whether two one-ways would have cost the same anyway, because domestically they usually do.

A worked example, start to finish

Concretely, here is the whole system applied to one trip. A traveler in Chicago wants ten days in Europe, sometime in autumn, and has about $500 to spend on flights.

1. They do not search “Chicago to Paris”. They open the Explore map from ORD with a flexible ten-day window across September to November, and add MKE as a second origin — 90 minutes north, and priced as a leisure market rather than a hub.
2. The map reorders their preferences. Paris is $780. Lisbon is $412. Dublin is $398. Milan is $445. The trip they wanted and the trip that is on sale turn out to be different trips, and they did not have to give up anything real to notice.
3. They check the price graph. Dublin has run between $850 and $1,050 for months. $398 is not a low number; it is an anomalous one. That distinction is what makes it worth acting on.
4. They set alerts and stop looking. Three routes, tracked. Nothing more to do for a few weeks.
5. An alert arrives on a Tuesday morning. Dublin, $398 round trip, travel windows in late October and early November, one stop each way, on a carrier they have heard of.
6. They use the 24-hour rule as a hold. Book it, confirm the time off that evening, keep it. If the answer had been no, they would have cancelled for free the next morning.

Total time spent: about twenty minutes across six weeks. Difference against the Paris fare they started with: $382. The mechanism was not a trick. It was refusing to fix the destination before seeing the prices.

The twelve-point checklist

Everything above, compressed into the order you would actually do it.

1. Decide the shape of the trip — region, rough season, length.
2. List every airport within two hours of home. Search all of them.
3. Search with flexible dates first, fixed dates never.
4. Use the Explore map to let price choose the destination.
5. Check the price graph to learn what normal looks like for that route.
6. Know your region’s booking window and start watching before it opens.
7. Set alerts on your two or three real candidates.
8. When an alert arrives, compare it to normal, not to your budget.
9. Price the fare on the airline’s own site before booking anywhere else.
10. Add bags and seats before you decide basic economy is cheaper.
11. Use the 24-hour rule to hold anything you need a few hours to think about.
12. Book direct, and book fast. Good fares are corrected, not discontinued.

None of this requires being a frequent flyer, and none of it requires spending your evenings on airfare. It requires accepting one thing that most travelers resist: the cheapest trip is rarely the trip you already had in mind. Loosen that, and the rest is mechanical.