Why subscription models and flash sales don’t work for airline fares – Travel101 Series

airplane fly in the sky photo licensed from Adobe Stock

Why can’t you just sign up to pay $X thousand a year and get all you can eat flights on your favourite airline? Why isn’t there a Netflix for city breaks? (For once it isn’t a legacy systems issue!)

Michael O’Leary was asked this question on Monday at the Future Travel Experience event in RDS, Dublin, and although he explained it well, he spoke quickly and assumed some level of awareness of revenue management. I’m going to try and explain it using more words and for a general audience. (If you are an airline revenue manager this might bore or annoy you – but if you can read it anyway and help me correct any oversimplifications I’d really appreciate it.)

You may have noticed that the prices of flights change all the time! What is happening is that the price is being controlled by the revenue management department and the software and algorithms they use. They combine historic booking data (this flight yesterday, last week, last year), and demand for this flight (the rate seats are selling), and other data sources, to adjust the price on the fly.

In principle, the higher the demand the higher the price. Also, the closer to the flight the higher the price (most of the time).

Another way of saying this is that the airline didn’t set the price – YOU – as the consumer – set the price. At least you acting along with everyone else who wanted to fly on that flight. Revenue management is like the simulation of an auction.

Airlines typically fill 8 out of 10 seats this way (80% is the industry average load factor), with some, like Ryanair, consistently filling more than 9 out of 10. So why don’t they then sell the last seats off really cheaply – in a last minute flash sale – so they always have 100%? Well, that would mean they aren’t making the maximum money possible (optimising yield) from each flight.

Imagine you are an airline and you have only 8 seats left on a flight and 6 hours to go before take off. If your data forecasts a strong chance someone will show up and pay €500 then you don’t just sell them off for cheap. The yield optimisation algorithms have worked out that €500 is the best bet – sell at €600 and you’ll drive people away, sell at €400 and you won’t sell more seats you’ll just make less money overall. (This is based on “elasticity of demand”.)

Another reason not to do last minute price drops is that they train consumers to wait until the last minute to get bargains in the future, which would hurt airlines a lot in the long run. So the big airline sales you see aren’t really about distressed inventory, normally they are being done far in advance by the sales department to stimulate demand, and invariably they are fixed to limited number of seats (by revenue management). Norwegian’s €69.99 to the USA from Europe is not an average fare, it’s their best possible fare. For now, Norwegian have very little historic demand data to go on.

Airlines call maximising the money from each flight “yield optimisation”, and it is an activity usually carried out by the revenue management team. I’ll do a blog post (or a video) explaining revenue management in more detail at a later date, maybe with help from an actual revenue manager, for now, though back to subscriptions.

The core problem with a subscription is that they remove the ability of the airline to maximise yield (day by day) as demand changes.

The airline will be able to sell subscriptions – that is not the issue – and they would get some loyalty and cash forward benefits. Let’s consider two user scenarios:

Heavy User

A frequent flyer (business), who has flown (for example) 10 times last year and wants to buy the next 10 tickets for a fixed price, will absolutely pay in advance if she sees some saving or another benefit. But the airline doesn’t need to sell her a subscription, she will buy the flights anyway. And those frequent business travellers pay a premium for flexibility and booking at the last minute. Giving them a discount is taking away from your best source of revenue. (What marketing and revenue managers call “cannibalism”, when one of your products eats the another’s market).

Light User

On the other hand, an infrequent traveller, who is flexible about their times and dates (willing to travel mid-day or mid-week for a cheaper fare for example), is going to be a tough sell for a subscription. He figures another fare sale will come along with €9.99 fares and he’ll grab a ticket then. The price he would be willing to pay for a subscription for 10 flights would be dangerously low for the airline, especially if he shows up expecting to board a flight that should now be selling at €500 seat for last minute bookings.

Is there a way to keep subscriptions benefits and somehow still maximise yield? A subscription essentially fixes the price for one consumer, they don’t need to select flights based on price.

Another way of asking this question is – when a passenger has bought a subscription what limitations can the airline reasonably place on it so they can still control yield?

  • Do you force everyone to specify their flight X days in advance?
    This is a sensible way to pay less for fares anyway, and it doesn’t hep the business traveller who needs to make a lot of last minute changes.
  • Do you limit to specific max price or fare class?
    Firstly “fare class” are a thing most consumers have no real awareness of, and secondly, you are asking someone to pay up front but also take all the risk on prices going up and having the pay the difference, most people won’t go for that.

Many such limitations or variations on the subscription product have been considered by airlines over the years. But the limitations necessary to not undermine yield optimisation, would invariably undermine the consumer value of having a subscription in the first place.

In the end, airlines can be weird, and there’s no such thing as an absolute rule – as soon as some consultant says “no airline will ever do X”, an airline does it. I’m not trying to be that consultant.

In airlines and in travel generally, subscriptions for bundled products or extras do exist, as do subscriptions for percentage discounts and other benefits – but often the core fare still varies flight by flight. Airports and lounges find subscription models work very well – they provide convenience benefits for frequent users and help fix travel budgets in advance, but they operate on (mostly) fixed price models in the first place. Paid membership programs for some frequent flyer benefits are out there also, but they are not common.

In general, airlines are getting more dynamic with their pricing. They are finding ways to do it more quickly and effectively, to make shopping on the airline website or sites like Skyscanner more accurate. Also, a lot of things which are fixed price today, like baggage fees, may become dynamically priced in the future. This will make the introduction of subscription products and last minute sales even more difficult.

This is the first of (hopefully) many Travel 101 posts, designed for people relatively new to the industry. The content will suplement the Travel 101 training program. If this post helped you understand something about the airline industry, please share it with others, or let me know. I don’t know if I should keep writing stuff like this, so tell me if you want more? Any feedback or corrections welcome, I’ll advise of any updates below. Thanks for reading, Mark Lenahan. [This post originally appeared on marklenahan.com.]

2 thoughts on “Why subscription models and flash sales don’t work for airline fares – Travel101 Series

  1. thewanderingaramean

    The missing part of the query/response at FTE was the value proposition of such offers outside the revenue management system. It was asked in the context of JetBlue’s All You Can Jet offering, a flat-rate play offering unlimited flights for 30 days in off-peak times. I did it twice.

    But JetBlue wasn’t doing it to maximize revenue on those flights. It was a marketing effort to fill seats that RM had already predicted would be empty. It was also sold in very limited quantity so the chance of eroding the higher yield business travel demand was relatively low. And the other induced travel spend was all positive. MOL even commented about higher demand travel dates as a challenge; in the second iteration JetBlue had two prices, one that included Friday/Sunday travel and one that excluded it.

    Offering a year-long option would be foolish for an airline. But for 30 days it can be a great marketing tool, once you get over the scare that it is going to trash yields. There are ways to control for that if the airline wants to.

  2. Valentin Dombrovsky

    Great article, Mark, thank you for sharing that.

    But I personally see that in the case of “heavy user” we propose that we have a frequent flyer who is using only one airline and doesn’t look at the competitors. But what if subscription could make him or her use your airline for all 10 flights, thus helping to fill in the seats and to earn a bit more.

    Also, ancillary sales are left out from the whole picture. Having more seats sold is not only the matter of money made on airline tickets, but also additional revenue opportunity.

Comments are closed.