On the 17 routes where we had enough data to compare airlines fairly, the typical gap between the cheapest and the most expensive carrier on the same route was 17%. On 8 of those 17 routes it was 20% or more. Same route, same period, same economy cabin — a fifth of the ticket price, decided by which airline’s name you clicked.
That is the finding. The rest of this piece is how we measured it, what the exceptions look like, and what it means when you are actually booking.
What we measured
Between June and September 2026 the scanner behind theflightagent.app logged 21,675 one-way economy fares while watching routes on behalf of real travellers. Every fare is the cheapest the search returned at that moment, for one adult, in economy.
For each route we took the median fare per airline and compared the cheapest airline against the dearest. Medians, not minimums: one freak quote is a story about a single day, not about an airline. We kept only routes where at least two airlines each appeared 40 times or more, which left 17 routes.
The results
| Route | Cheapest airline | Most expensive | Gap |
|---|---|---|---|
| Madrid to Beijing | Beijing Capital Airlines ($325) | EgyptAir ($575) | 44% |
| Marseille to Bali | China Eastern ($600) | Swiss ($873) | 31% |
| Washington to Chicago | Frontier Airlines ($92) | United Airlines ($133) | 31% |
| Dubai to Chicago | Royal Jordanian ($521) | EgyptAir ($739) | 29% |
| Paris to Shanghai | Hainan Airlines ($442) | Aegean Airlines ($615) | 28% |
| Vienna to Tokyo | China Eastern ($532) | Etihad Airways ($744) | 28% |
| Winnipeg to Las Vegas | United Airlines ($151) | Air Canada ($192) | 21% |
| Porto to Cairo | LOT Polish Airlines ($242) | Aegean Airlines ($307) | 21% |
| Luxembourg to Shanghai | Luxair ($468) | Air India ($564) | 17% |
| Marseille to Jeddah | flynas ($307) | Pegasus Airlines ($355) | 13% |
| Toronto to Beirut | TAROM ($591) | Turkish Airlines ($675) | 12% |
| Tel Aviv to New York | HiSky ($597) | Neos ($679) | 12% |
| Bengaluru to Toronto | Delta Air Lines ($624) | Air Canada ($688) | 9% |
| Madrid to Shanghai | Hainan Airlines ($357) | Brussels Airlines ($392) | 9% |
| Tokyo to Toronto | China Eastern ($771) | EVA Air ($849) | 9% |
| Sydney to Hyderabad | A1 ($435) | Cathay Pacific ($472) | 8% |
| London to Bangkok | China Southern ($410) | Air China ($416) | 2% |
The widest gap was on Madrid to Beijing, where Beijing Capital Airlines sat 44% below EgyptAir. The narrowest, London to Bangkok, was effectively nothing.
How much the threshold matters
A study like this lives or dies on an arbitrary choice: how many times must an airline appear before we are willing to compare it? Set the bar too low and a carrier seen four times can manufacture a dramatic gap. Set it too high and you throw away real routes. So here is the same analysis at three bars.
| Minimum observations | Routes kept | Median gap | Routes at 20%+ |
|---|---|---|---|
| At least 25 fares per airline | 20 | 18% | 9 |
| At least 40 fares per airline | 17 | 17% | 8 |
| At least 50 fares per airline | 17 | 13% | 7 |
We report the middle row. The headline number moves between 13% and 18% depending on where the bar sits, which is worth knowing before anyone quotes it as a single fact. What does not move is the direction: at every threshold we tried, a large minority of routes showed a gap of a fifth or more.
The cheapest airline is rarely the one you thought of first
Look down the left column of that table and the pattern is hard to miss. Beijing Capital Airlines. Royal Jordanian. HiSky. AJet. flynas. TAROM. These are not the names that come to mind when you picture a route.
Now look at the right column: Swiss, United, Etihad, Air Canada, Aegean, EgyptAir. Larger, better known, and on these routes, more expensive.
We want to be careful about how strong a claim we make here, because it is the kind of thing that is easy to overstate. We did not find that the airline based at one end of a route is reliably the dearest — the expensive column contains plenty of carriers with no hub at either end. What we found is narrower and more useful: the cheapest option on a route is usually a carrier you would not have searched for by name. A default search that surfaces the familiar airlines is showing you the expensive half of the market.
Three genuine exceptions
The rule is a prior to check, not a law, and the exceptions are instructive.
Luxembourg to Shanghai. Luxair, the home carrier at the departure airport, was the cheapest airline on the route. Luxembourg is a small market where the local airline feeds passengers onto long-haul partners rather than defending a monopoly.
Madrid to Beijing. Iberia flies it non-stop and priced within about $20 of the cheapest one-stop connection. When the home carrier has real scale on a route, the convenience can come close to free.
London to Bangkok. A 2% gap. Some routes are simply competitive enough that no carrier can hold a premium, and on those this whole exercise is a waste of your time.
What to do with this
The practical version is one sentence: before accepting the fare from the airline you first thought of, spend a minute pricing the unfamiliar names on the same dates.
Two things to weigh against the saving:
- A long connection is a real cost. Much of this gap is the price of a shorter journey. If the cheap fare turns a day into an overnight, or lands you somewhere at 3am, it may not be cheaper in any sense you care about.
- Compare what is included, not just the fare. Baggage allowances differ sharply between full-service and low-cost carriers, and a stripped fare plus two bags can beat a bundled one or lose to it. Confirm on each airline’s own site for your exact ticket — that is the only place those numbers are authoritative.
Limits of this analysis
We would rather state these than have them found for us.
- These are one-way, economy, one-adult fares, and the cheapest available at the moment of each scan. Return and premium-cabin pricing may behave differently.
- The routes are the ones our users asked us to watch. That is a real-world sample rather than a designed one, and it skews towards Europe, the Gulf and long-haul Asia.
- We compare each airline’s median across the whole window rather than matched departures. An airline that happened to serve cheaper dates would look cheaper partly for that reason — though the effect would have to be very large to explain the gaps at the top of the table.
- We publish no “book N days ahead” figure, here or anywhere. Our scanning pattern observes most flights on a single day, so any lead-time curve we drew would compare different flights at different lead times and present route mix as a trend. It is the single most requested number in this field and we think most published versions of it are measuring the wrong thing.
If you would rather have this done for you, theflightagent.app watches every carrier on a route at once and emails you when any of them drops below the price you set — which is, in effect, this analysis run continuously on the one route you care about.
Frequently asked questions
How much cheaper is one airline than another on the same route? In our data the typical gap between the cheapest and dearest carrier on a route was 17%, and on 8 of 17 routes it was 20% or more.
Is the national flag carrier always the most expensive? No, and we specifically did not find that. Luxair was the cheapest airline on Luxembourg to Shanghai, and Iberia priced its Madrid to Beijing non-stop within about $20 of the cheapest connection. The reliable pattern is that the cheapest carrier is an unfamiliar name, not that the familiar one is always dearest.
What was the biggest price gap you found? Madrid to Beijing, at 44% between Beijing Capital Airlines and EgyptAir.
How many fares is this based on? 21,675 one-way economy fares logged between June and September 2026, filtered to 17 routes where at least two airlines each appeared 40 times or more.
Why do you not publish a best day to book? Because we cannot measure it honestly with this data. The same flight is usually observed on a single scan day, so a lead-time curve would compare different flights at different lead times and present route mix as a trend.
Fares and schedules change constantly. The price data here was collected up to 20 September 2026 and last checked in September 2026; always confirm on the airline’s own site before you book.
Comments