Nobody wants to find out their plane ticket cost twice as much as the person next to them. Buying your ticket at the right time can save you upwards of 40%.
Getting the best price on your air travel is a confusing and frustrating proposition. Flight prices are less transparent and fluctuate more than almost anything else a consumer buys, even though airfare is one of the most expensive purchases that a typical family makes.
Just think: buying a car might cost you more, but its price is much less mysterious. Although the price of gas for that car does vary, you won't end up spending three times more than your neighbor just because you went to the wrong gas station at the wrong time.
Airfare prices change all the time: airlines change their prices on a daily or even hourly basis in response to market conditions, use demand-based “yield management" to control which fares are available on a given flight, and launch unpredictable “flash sales" to grab market share. Figure 1 shows how much prices vary even for just one airline's nonstop flights on a specific date.
All this causes plenty of fear, uncertainty and doubt: consumers typically spend almost two weeks comparison shopping and end up spending nearly 5% more than when they started looking. Even after all the comparison shopping, they're not confident they've got the best deal, and end up buying almost out of a sense of frustration with the process.
Ironically, all the variability in prices means there are good deals to be found. Consumers lose out because they're only spot-checking prices occasionally: we found that by watching continuously there's often a price 5-10% lower than the first price even within 24 hours.
Figure 1: Actual prices quoted over time for nonstop flights on a single airline departing June 1 from Boston to San Diego, and returning on variety of dates (color-coded). Note the high degree of price variation, as well visible “steps" in pricing, triggered by fare-class advance eligibility restrictions.
Our goal at Hopper is to bring more transparency to pricing, by giving consumers advice about where and when to fly -- and when to buy -- to save money on their air travel. We believe this helps consumers buy with more confidence, more quickly which leads to them being more satisfied with their ultimate purchase.
The Hopper app brings all our expertise to your mobile device, with advice about the best dates to travel, alternate airport choices, and even tips about which specific flights will serve you best. One of the key features is our “when to buy" advice. Once you've decided where you're headed, we'll watch prices for your trip continuously and send you an alert when we see a great deal. This is quick and easy for the consumer, but tricky to pull off.
The question is what makes a good deal? If we're too conservative and recommend that you buy too early, we risk missing out on a good deal later, but if we're too optimistic we might wait too long looking for a price that will never occur, and you'd end up paying more as prices spike upwards as your departure date approaches.
Because prices change in unpredictable ways, at the whim of the airlines, it's impossible to know for sure. But it turns out there are a number of predictable factors that we can use to make an educated estimate about how prices will move, saving the consumer about 20% on average, and upwards of 40% in some cases.
(At least until we tell you to!)
Once you've chosen where you're going, the major drivers of your ticket price are when you're traveling and how far in advance you book. Travel around popular dates (Thanksgiving, Christmas, school vacation weeks, etc) often costs significantly more, but in most cases is quite predictable by comparing prices now with prices at a similar point last year (figure 2).
Figure 2: Seasonal variation in good round-trip prices from Boston to San Diego, showing predictable variation with popular dates (school vacation weeks, summer break, Thanksgiving, Christmas/New Year).
When you've picked your travel dates, the remaining question is how far in advance to book your ticket. Most of the variation in ticket price is driven by the yield management systems of the airlines. They constantly adjust prices up and down based on how quickly flights are filling up, with the goal of selling each seat at the highest possible price while still filling (or slightly overfilling!) the plane before it takes off.
A major challenge for the airlines is that the people willing to pay the most for tickets, like business travelers, don't know that they need the ticket until the last minute. So airlines need to trade off selling cheaper tickets early on to consumers against holding back enough seats to sell at higher prices later on.
This leads to a fairly predictable “advance purchase" curve: airlines set an initial price that's slightly high to capture unanticipated demand along with people looking to lock in specific dates far in advance, then the yield systems push prices down slightly until demand stabilizes and prices bounce around some central trend, and finally prices start rising rapidly closer to departure as the less flexible and less price-sensitive business travelers enter the market.
Every market looks different, for example prices often don't vary as much or rise as quickly in markets without a lot of business travel (see figure 3). Using historical data, we analyze how prices are likely to trend for your destination and dates, and how much fluctuation to expect. This lets us give you recommendations about the best time to book, on average, as well as how long you can afford to procrastinate.
Figure 3: Business-oriented routes like Newark to Boston (left) tend to have low, slightly declining prices at long advances, with prices and demand then rising very rapidly prior to departure. More consumer-oriented markets like Newark to Honolulu (right) are booked much farther in advance, and have a much smaller last-minute premium. By analyzing these price trends, we can tell you how far in advance to book.
If prices tend to rise as you get closer to departure, why would you ever wait to buy your ticket? The answer is volatility: a fancy word that measures how much prices fluctuate around their central trend. When there's a lot of volatility, the actual price will often drop compared to today's even when the trend is going up (see figure 4).
Figure 4: Price volatility means you can often find a cheaper price even when the overall price trend is rising: the first red dot is priced about $345 (above trend) but later dips under $310 (below trend). It also illustrates why you tend to pay more if you procrastinate and only check prices once in a while - since the trend keeps rising - but can still often save money by waiting if you're watching prices continuously.
The trick is to decide whether it's worth not buying the ticket today and instead waiting in the hope of seeing a lower price in the future. That's where the rocket science comes in. We study historical pricing patterns in every market to understand how much and how often prices tend to change. This lets us estimate how much prices are likely to move around the central trend, and how likely it is you'll find a significantly better price than the one you see today.
Again, it depends on the market. Based on the central trends shown in Figure 3, it looks like it would be worth waiting until about a month ahead for the Newark to Boston ticket, and in contrast that you should buy the Newark to Hawaii ticket as early as possible. But in fact there is much less price volatility for Newark-Boston compared to Newark-Hawaii. Our calculations show that if you start searching for Boston at least two weeks in advance it's only better to wait about a third of the time, and even perfect buy or wait decisions would only save about $7 over your initial price. On the other hand, for Newark to Honolulu, even though the trend is consistently rising it's better to wait about 83% of the time, with perfect buy or wait decisions saving you almost $90 over the initial price.
Obviously, we can't make predict the future perfectly: there's always a risk that prices don't do what we expect, and sometimes you just plain get unlucky. If you absolutely, positively must have a ticket for specific dates you should buy as soon as you see a price you're comfortable with. However, we've back tested our recommendations and in 95% of cases, following the advice from the price prediction will get you a cheaper flight or at least the same price as your initial search.
With the standard proviso that past performance is no guarantee of future success, we analyzed data from billions of trips in our historical archive to measure how our predictions would have performed for those trips.
We found that about a third of the time the price you see in your first search is the best it's going to get, but more often—about two-thirds of the time—the price for your trip will be lower at some future point before departure. The earlier you start watching, the more you can potentially save. Figure 5 shows that if you start watching a trip at least six months out, the typical saving from the first price you see and the best future price is about 24%. If you start watching two weeks ahead, the best you can hope for is saving about 7%.
Figure 5: If you start watching at least 180d before, then the potential savings between the median price when you start watching, and the best future price, is about 24%. If you start watching two weeks ahead, the potential is only 7%.
Our recommendations aren't perfect, since we have to make a decision about today's price without knowing for sure what the best future price is going to be, but on average following our buy or wait recommendations will save you about 20% of your ticket price.
As with everything to do with airfare, there is plenty of variation between markets (where you're flying to and from), and between different dates within markets. As shown in table 1 and 2, we found that for some popular destinations, like Orlando or Las Vegas, or consumer favorites farther afield like Montego Bay, Aruba, Paris, or Rome, you can regularly save more than 20% by watching, and on some routes, you'll save upwards of 40% over the first price you're quoted!
Table 1: Potential savings for domestic destinations from a “typical" US origin.
Table 2: Potential savings for international destinations from a “typical" US origin.