AiTravel.deals does not send price alerts. There is no alert form and no mailing list — the destination pages here re-price every hour, and that is the whole product. This guide is about the tools that do send alerts, how to read what they tell you, and where they let you down.
- Google Flights alerts are free, cover essentially every route, and take about a minute to set.
- Save the alert against a date range, not two fixed days. This is the setting people get wrong.
- An alert tells you a price moved. It cannot tell you whether the price is good — that is a different question, and the one you usually have.
- Inside three weeks of departure, fares mostly move one way. An alert at that range is a notification service for bad news.
Most people book flights by opening a search, looking at a number, and having no idea whether that number is good. A price alert is supposed to fix this. It half does. It will reliably tell you that a fare moved. It will not tell you whether the fare was worth having in the first place, and the gap between those two things is where most of the disappointment lives.
Here is what the tools actually do, how to set them so they are worth reading, and the cases where an alert is the wrong instrument entirely.
What a price alert actually is
A price alert is a saved search that something re-runs on a schedule. When the result comes back materially different from last time, you get an email. That is the entire mechanism. There is no negotiation, no inside knowledge of what the airline intends to do next, and — in the free tools — no serious attempt to tell you whether the new number is objectively low.
What you specify is usually:
- A departure city, or a set of nearby airports
- A destination, or "anywhere" on the tools that support it
- A travel window — the setting that matters most, and the one people get wrong
The alert then fires on movement, not on value. A route that falls from $1,400 to $1,150 will trigger one even if $780 is the ordinary price for that route in a normal month.
Setting one up in Google Flights
Google Flights is the one to start with: free, near-universal route coverage, and the alerts arrive quickly. You need a Google account and about a minute.
Search the route, not the date
Enter origin and destination. For dates, open the calendar and use the flexible-dates grid rather than picking two specific days — it shows what each departure costs, which is information you want before you commit to a window.
Widen the window before you save
Pick a range of a fortnight or more, or switch to the mode that lets you say "1 week in September". An alert saved against two fixed days can only ever report on those two days, which throws away most of what makes fares move.
Turn on "Track prices"
The toggle sits under the search bar on the results page. On many routes there is a second switch for tracking any dates rather than only the ones you chose — worth enabling, since it widens the alert to the whole calendar.
Add neighbouring airports as separate alerts
If you would drive ninety minutes to save $200, save a second alert from that airport too. Fares move independently between hubs, and a tracked search will not do this for you unless you searched a multi-airport origin in the first place.
Everything lives under "Tracked flights" at google.com/travel/flights. Alerts do not expire, so prune the list occasionally — past a certain number you stop opening the emails, and an unread alert is worth nothing.
What the "prices are currently low" banner means
Google shows a coloured banner on some routes saying prices are low, typical, or high. It is genuinely useful and widely misread. It compares against the historical range for that route and those dates. It is not a prediction, and it is not a claim that you cannot do better.
The banner judges the search you ran. Shift your departure by four days and you may get a different fare with a different banner. "Prices are currently typical" on an expensive week is not the same information as "this is what the route costs".
Hopper and similar apps go further and offer a buy-or-wait recommendation. Treat these as a confidence signal rather than a forecast: the accuracy figures are self-reported, the model is not open to inspection, and the app has a commercial interest in you booking through it. Useful as one input among several. Not a reason to sit on a fare you are already happy with.
Five settings that make alerts worth reading
Track two or three origins
Deals land unevenly across hubs. Transatlantic fares from Boston and New York can differ by a couple of hundred dollars in the same week for no reason visible to a traveller.
Give it a fortnight, minimum
A two-week window catches the mid-week departures that carry most of the saving. A fixed Friday-to-Sunday alert will mostly tell you that weekends are expensive.
Keep one open-destination alert
Google's explore map and Kiwi's anywhere search both take a budget instead of a place. Worth one saved search if your reason for travelling is "somewhere warm".
Check what the fare includes
A $60 drop that arrives with the cabin bag removed is not a drop. Alerts compare headline fares, and basic-economy repricing is a common source of phantom good news.
Stop tracking inside three weeks
Close to departure, fares mostly move in one direction. Keeping the alert on past that point buys you nothing but a running commentary.
Which routes reward tracking
Alerts pay off where a route has real volatility — several carriers competing, a wide seasonal spread, and enough capacity that a sale actually happens. They do very little on thin routes with one operator, where the fare is simply what it is.
- Britain to North America: the deepest competition in any long-haul market and the widest seasonal swing. The gap between February and July on the same route is the largest single saving a flexible date will buy you.
- Britain to South East Asia: high capacity through the Gulf hubs, recurring fare wars between them, and a shoulder season most travellers ignore.
- Britain to the Canaries, Morocco and Egypt: short-haul, heavily promoted, sold hardest against the package holiday, and prone to flash sales a tracked search will catch.
- One carrier, one daily flight: skip it. There is nothing for an alert to detect.
Where alerts stop being the right tool
An alert answers "has this changed?". It does not answer "is this a good price?" — and on a route you have never flown, the second question is the one you actually have.
Answering it means comparing against what the same route is charging other people right now: the median of what a search returns, not the cheapest line in it. That is the number that tells you whether $640 to Lisbon is a find or an ordinary Tuesday. It is also what the destination pages on this site are for — each one re-prices hourly and shows the fare beside the median of thirty live fares pulled on the same route at the same moment. The method is written up here, including where it is deliberately conservative and what it cannot tell you.
Common mistakes
- One alert, one date pair. The single biggest cause of alerts that never fire. Widen the window.
- Treating the alert as a verdict. A drop from a high starting point is still a high price. Establish the route's ordinary level before acting on movement.
- Twenty tracked searches. Past about five you stop reading them.
- Waiting for the floor. There isn't one. If a fare is clearly under what the route usually costs and the dates work, holding out has negative expected value.
The short version
Set two or three Google Flights alerts on wide date ranges, from every airport you would realistically use, three to six months out. Read the banner as context rather than instruction. And when one fires, spend thirty seconds working out what the route normally costs before deciding the number in the email is good news.