How to Find Cheap Flights You'd Actually Take (Not the Incognito-Mode Myth)
Most flight-saving advice repeats the same handful of tips without checking whether they actually work. A few of them do. Most don't. Here's what's genuinely moving the price, and what's just superstition. And a cheap flight is only worth having if it's one you'd actually get on: this is about finding fares that are properly cheap and properly workable, not just the lowest number on the page.
The myth worth killing first
Incognito mode doesn't lower flight prices. Neither does clearing your cookies, using a different device, or booking at 3am on a Tuesday because someone read that airlines "reset" prices then. Airline fares are driven by seat inventory and demand algorithms, not your browsing history. There's no evidence any airline has ever priced a ticket based on whether you've searched for it before. If a price changes between two searches minutes apart, it's because the fare class sold out or demand shifted, not because you got caught looking twice.
What actually works: date flexibility
The single biggest lever on any route is which exact days you fly, not which airline or booking site. The same route, same cabin, same week can vary by 2-3x depending on whether you fly Tuesday or Saturday, and whether you're in shoulder season or peak. This is well established and it isn't subtle: airlines price by demand for that specific departure, and demand for a Friday evening flight is nothing like demand for a Tuesday morning one.
The problem is that checking this properly means comparing dozens of date combinations across a window, not just the one weekend you had in mind. Most people check two or three dates and call it done, because checking forty by hand is tedious.
This is exactly the gap tools exist to close. Google Flights' own date grid and Momondo's calendar view will show you a whole month at a glance instead of one fixed pair, and a milo alert takes it further by scanning a broad date window for you on an ongoing basis and surfacing the cheapest match it finds, so the checking happens automatically instead of forty tabs deep in a browser.
What actually works: waiting and watching, not booking the first price you see
Prices for the exact same route and dates move day to day, sometimes by a meaningful margin, even when nothing about the flight has changed. Booking the moment you start looking means booking blind to whether that day's price is typical or unusually high. Knowing roughly what a route normally costs, and having a target price in mind before you start, turns "is this a good deal" from a guess into a fact you can check.
Working out what "normal" looks like for one specific route is the harder part, and it's worth saying plainly: asking a general-purpose AI chatbot won't help here. It has no live feed of what this route is trading at this week, so whatever number it gives you is a plausible-sounding guess, not data. What answers that question is Hopper's watch-and-predict feature, which will tell you whether it thinks a fare is likely to rise or fall, or milo's own price analytics, which show you where today's price sits against that route's recent history, so the decision to book now or hold off is based on something real.
What actually works: a bit of flexibility on stops and airports
Accepting one stop instead of insisting on direct, or checking a nearby airport as well as your usual one, regularly opens up fares that a rigid search never sees. It's not glamorous advice, but it's real.
The usual objection is fair, though: nobody agrees to "one stop" wanting to find out it means an overnight stranding in a departure lounge they've never heard of. That's a genuine risk of blind flexibility, not a reason to rule stops out altogether. The fix is being flexible on the concept rather than the specifics: setting a maximum layover time, or ruling out particular connecting airports, means you only ever see the stopovers you'd actually consider, not every technically-cheaper routing regardless of how miserable it'd be to sit through. More on why that distinction matters in The Cheap Flight That's Actually Worth Booking.
Why most people who do this properly end up automating it
Doing all of the above by hand, for one trip, is genuinely a few hours of work: checking dozens of date pairs, tracking a route's typical price over time, watching for the day it dips. Doing it for every trip you might take is not realistic for anyone with a job. That's the actual reason tools like milo exist: not to replace judgement about what a good deal looks like, but to do the repetitive checking that a human reasonably won't keep doing by hand, and flag it when a route hits a price worth acting on.
None of this guarantees the cheapest possible fare every time. Nothing does. But it's the difference between guessing and knowing whether a price is good, on a flight you'd actually get on rather than just the technically-lowest number on the page.
For more on setting up an alert across a whole window of dates rather than one fixed pair, see Flight Alerts for Flexible Travel Dates.