Do flight prices really rise when you search twice?
The belief is close to universal. The evidence for it is thin, and the mechanism that actually moves the price is both simpler and less personal.
Almost everyone believes it: search for a flight, come back later, and the price has gone up because the airline knows you are interested.
It is a satisfying explanation. It is also, on the available evidence, mostly wrong — and the real mechanism is worth understanding, because it changes what you can actually do about the price.
What actually sets the number
Airline seats are not sold at one price. They are divided into fare buckets — blocks of seats at different prices on the same flight, in the same cabin, on the same day.
The cheapest bucket holds a small number of seats. When those sell, the system sells from the next bucket up. Nothing about the flight has changed, and no decision has been made about you. The cheap seats are simply gone.
This repricing runs continuously, driven by how fast seats are selling relative to how fast the airline expected them to sell, across an enormous number of flights.
Why the belief persists
Because the experience is real. People genuinely do see a higher number on the second look. Three things explain most of those cases.
A bucket emptied between your two searches. On a popular route this can happen in minutes. The change is real, permanent, and has nothing to do with you.
The first result was cached. Search tools hold results briefly. The first number you saw may already have been minutes old. The second search returned live data — which looks like an increase and is actually a correction.
You took a different path. A different site, different dates, or different filters can return a genuinely different fare product — different bucket, different baggage allowance, different refundability — displayed as though it were the same seat.
What researchers have found
People have looked for this, repeatedly, and the effect is hard to find.
- A 2015 review by the US Department of Transportation, prompted by complaints about discriminatory pricing, found no conclusive evidence that airlines set fares from browsing history.
- A 2020 Consumer Reports test compared prices across different browsers and private-browsing modes, and concluded the variation it saw came from timing and system updates rather than tracking.
- Academic work examining the three largest US carriers likewise found no evidence of price discrimination based on browser cookies.
Where differences did turn up in these tests, they were usually explained by the currency and region being served, or by which fare product was being shown — not by who was doing the searching.
That is not the same as proving nothing of the kind exists anywhere, and “surveillance pricing” is an active area of regulatory attention in other industries. It does mean the confident version of the airfare story — they raised it because they saw you looking — is not well supported by the people who have gone looking for it.
The incentive is worth noting too. Repeatedly quoting a higher price to an interested buyer who can open another tab is a poor strategy in a market where the identical seat is sold through many channels at once.
What genuinely moves the price
- How far ahead you book. The strongest single variable, though the shape of the curve differs by route and season
- How full the flight already is. Demand, not your interest in it
- Season, holidays and events. Known in advance and priced accordingly
- Which fare product you are comparing. The cheapest result on two sites is often not the same thing
What this means in practice
Not worth your effort: clearing cookies, private browsing, pretending to be in another country, or searching only on particular weekdays.
Worth your effort:
- Check the same flight on the airline’s own site before assuming a difference is real
- Compare like with like — bag included or not, changeable or not
- Set a price alert rather than searching repeatedly, so you hear about it when it actually moves
- Book when the price is acceptable, rather than waiting for a floor you cannot identify in advance
The useful reframing
The price did not go up because you looked. It went up because somebody bought.
That is less sinister and more actionable: you are not being watched, you are in a queue. And the only thing you control in a queue is when you join it.
Common questions
- Does clearing cookies or using private browsing get cheaper flights?
- There is little evidence that it does. Studies looking for personalised airfare pricing have generally found that the differences people notice are better explained by seat inventory changing, by different search paths returning different fare products, or by currency and region settings.
- Why did the price change between my first and second search?
- Most often because the cheapest remaining seats in that fare bucket were sold, or because a cached result from the first search had gone stale. Airfares are repriced constantly and independently of who is looking.
- Is there a best day of the week to book?
- Older advice about specific days comes from an era of weekly manual fare filing. Pricing is now continuous, so the reliable variables are how far ahead you book and how full the flight is, not the day you happen to search.