4 types of flight price alert: which you need

There are four kinds of flight price alert. They share a name and they are four separate products, each built to answer a different question.

Only one of the four judges every fare it finds against what that route normally costs, and keeps doing it across months you have not chosen yet. That kind is called a route watcher, and Flydar is one. If you have no dates, it is the only category that can take your trip as input at all. If you do have dates and they are months away, it is still the one that tells you a fare is low rather than that a fare changed.

The other three are useful, and each beats us at something named below. Here is the whole category in one view.

The four, side by side

Which one fits your trip

The decision mostly comes down to one thing: whether you want a tool that reports movement or a tool that judges the fare. Three situations cover almost everybody.

You know the route but not the dates. Nothing free can accept that as input. Every free alert is a saved search underneath, and a search needs dates, so you end up inventing dates and watching a guess. A route watcher takes the route, a trip length as a range, and a window of months. That is the entire input.

You know the route and the dates. People assume this case is already covered, and it is only covered if you are booking soon. A fare eight months out has a long way left to move. Free tracking will tell you that it moved: up $9 on Tuesday, down $40 on Friday, up again next week. It will not tell you that today’s number has finally landed below what that route normally costs, which is the only message you wanted. Flydar takes a fixed window as happily as a rolling one, and judges the fare instead of reporting it.

You have a shortlist rather than a single destination. “Somewhere warm in February.” “Japan or Korea.” “Anywhere in Europe in spring.” This is where people wrongly rule themselves out, because a shortlist is just several routes and you can watch several at once. Name a city and it stands in for every airport in it, at both ends. Three or four watched routes cover most of what “anywhere” turns out to mean once you say it out loud, and there is no supported-airport list to qualify for: if a plane flies it, it can be watched.

One thing nobody can tell you is when a drop will land. It might be this week, it might be in two months, and it does not announce itself in advance. That is the argument for keeping something pointed at the route continuously, not a reason to expect a long wait before anything happens.

1. Route watchers, and what Flydar does

What you name: a route, roughly how long you would go for, and a window of months. What it decides: whether a fare is low for that route.

This category inverts what the other three take as fixed. There is no saved search, so nothing can expire. What you hand over is a description of a trip rather than a query, and that one change is what lets something watch on your behalf for as long as you want it watched.

Four things follow from it, and they are the product:

• A city means every airport in it, at both ends, without you running six searches. Gaps of $150 between two fields 40 miles apart are ordinary, and this is how you stop losing them.
• Trip length goes in as a range. Fare rules have carried minimum and maximum stay conditions for decades, so the same route on the same departure date prices differently for a five-night trip than a nine-night one. “Exactly seven nights” and “six to nine nights” are two different requests, and the second one finds far more.
• The window can be as wide as fares exist for, which is around eleven months. Set it once. Nothing to rebuild in three months, nothing to quietly stop working.
• The judgement happens before the email is sent. A fare has to fall clearly below what that route normally costs to be worth your attention. Prices rising send nothing. A sale banner over an ordinary fare sends nothing. The ten minutes of research a raw alert leaves you to do, working out whether this number is good for this route, has already happened by the time anything reaches you.

Flydar tracks economy fares and books nothing, so you buy the fare where it lives. There is one plan with no tiers to work out, and you describe a route in a sentence rather than maintaining a spreadsheet of saved searches.

Where it is the wrong purchase: you are flying inside the next fortnight, when the cheap buckets are usually gone; you need a cabin above economy; or you want to be told where to go rather than to name it yourself.

2. Per-search alerts

What you name: an origin, a destination, and dates. What it decides: whether the price of that search has moved.

These are the free ones: Google Flights, Kayak, Skyscanner. You run a search, toggle tracking, and the tool watches the query you ran. Target-price alerts are a variant of the same thing with a threshold you supply, not a fifth category.

Google’s own documentation is clear about the trigger: among the notifications it sends are ones warning that prices for a tracked route are likely to go up, with an estimate and a confidence level attached (Track flights and prices). That is a deliberate feature, and it also means a “$9 increase” email and a “$240 drop” email land in the same shape from the same sender.

Where it beats Flydar: a trip you have decided on, with dates you know, that you are booking within the next few weeks. Over a window that short there is little room left for the fare to move and the noise barely bites, so paying anyone for it is waste. The full line between the two is worth walking before you spend anything.

Where it stops: the alert is welded to the dates you typed, so when they pass it stops meaning anything and nothing announces it. That is the most common reason an alert goes silent for months, and no vendor can patch it. It is what a saved search is.

3. Curated deal lists

What you name: your departure airports. What it decides: which destinations you hear about.

Going, Thrifty Traveler and Dollar Flight Club run this model. People and software hunt unusually low fares out of a set of supported airports and send the finds to the whole membership. You are buying discovery, and the service picks the destinations.

Where it beats Flydar: pure inspiration. If you will take a good fare to somewhere you had not considered, buy one of theirs. We watch what you name, so a service whose job is picking the destination for you is a different product rather than a feature we forgot.

Where it stops: you cannot request a route. If you have to be in Munich in April you are waiting for your route to appear in someone else’s newsletter, and it may never. Coverage is also a fixed list of departure airports weighted toward places with more routes, which is why Going’s own help centre carries an article titled “Why isn’t my home airport listed?”. That question exists because the answer is sometimes no, and a higher tier does not add your airport to the list. Prices are in the comparison of Going and its alternatives.

4. Price predictors

What you name: a specific itinerary, dates included. What it decides: whether to book it now or wait.

Hopper is the best known. It reads historical and current fares for the itinerary in front of you and returns a buy-or-wait call, often with an estimate of what waiting might save.

Where it beats Flydar: at the moment of the click. We can tell you a fare is low for its route. Whether it will be lower on Thursday is a call we decline to make, because getting it wrong would cost you money and cost us nothing.

Where it stops: a predictor presumes you already have dates, so it is a timing tool rather than something that watches a trip you have not planned. It speaks to the direction the number is heading, not to whether the number was reasonable to begin with. And the accuracy percentages that circulate for these tools come from the companies publishing them and have not been independently audited, so treat a confidence score as a claim rather than a measurement. The full breakdown of what a prediction can and cannot promise walks through Hopper’s own numbers alongside Google’s.

The short version

Four products, four questions: is this price moving, where should I go, should I click now, and is this fare low for this route.

The last one is the only question that pays, and it is the one you cannot keep answering by hand. Name the routes you would fly, with dates or without them, and leave your inbox alone until a fare drops clearly below what that route normally costs.

That is not a saved search. That is what Flydar is.