Revenue-Based Programs Aren’t King, They’re a Cop Out

Jay Sorenson has an interesting new report (.pdf) declaring ‘revenue is now king’ that revenue-based programs are the new ‘it’ in loyalty. The piece begins by citing all sorts of reasons why revenue matters to a company, which is of course both true and unenlightening.

While mileage alone doesn’t incentivize loyalty, the goal of a marketing program isn’t to ‘reward customers who spend the most’ it’s to move the needle on incremental business (that exceeds marginal cost). So while Sorenson is undoubtedly correct that mileage is an imperfect proxy for that at best —

Mileage flown is no longer a reliable proxy to represent the value of a customer. The era of deregulation and arrival of low cost carriers changed pricing stability; very often long distance flights in competitive markets were priced at a fraction of short hops. The business traveler paying $600 earned the same 900 FFP miles as the leisure traveler who nabbed a $75 fare. Airline managers questioned the fairness and effectiveness of methods which didn’t better recognize the value of premium customers.

It’s not the case that total revenue is a reliable proxy for profitability at the margin, either. Delta implemented a bold experiment in awarding miles based directly on ticket spend. United and American threw up their hands in a cop out by simply mimicking Delta.

Neither distance nor total revenue is an ideal proxy for profitability or incremental revenue contribution, which is what a loyalty program wants to drive. The current revenue models are ultimately a weigh station along a journey, but the monkey see, monkey do airline industry takes a long time to innovate.

Sorenson suggests that ‘all of the top 25 airlines use revenue to influence mileage accrual’ — which is true, premium fares earn a multiple of miles, in many cases discount fares earn a percentage of distance flown, with airlines around the world.

As Sorenson explains, “Every airline, except American, Delta, and United, currently uses the type of fare purchased by the consumer to help determine the final tally of a member’s mileage or point total.”

However the next statement — that “the distance flown by a member has lost much of its relevance” isn’t right both because distance is still the primary determinant of accrual (longer flights earn more than shorter ones in any given fare class) and because the claim that distance flown has lost relevance is a comparison over time and loyalty programs around the world using fare class multipliers isn’t new at all. European and Asian frequent flyer programs have always been more influenced by revenue than US airlines have.

The better question — the question that’s clarifying in a lot of situations — is what’s changed?

  • Delta adopted a direct revenue model for earning miles. They considered a revenue model for redemption, too (a chart with mileage costs corresponding to ticket price ranges) but customers in focus groups pushed back on the elimination of the ability to outsized value for their miles.

  • United simply mimicked Delta, just as they had done with minimum spend for status in in raising the spend requirement exactly as Delta did. There was some thoughtful internal opposition to the idea at MileagePlus, but Jeff Smisek personally preferred the revenue option.

  • American copied Delta as well. With the US Airways merger this was always going to happen in some form, US Airways had been working on a revenue-based program before Delta launched theirs. It was simply delayed by more urgent integration issues at the airline. Still, the AAdvantage team pushed back on the particulars, seeking to move towards a greater degree of emphasis on revenue in their own way. It’s ultimately Scott Kirby, now President of United, who pushed for simply precisely aping what Delta developed.

What’s changed is that the 3 largest US airlines have moved towards revenue-based accrual for flights (which remains a minority of miles earned, and is only half the earn-burn equation). Even that overstates the trend of course, other than to say that United and American do what Delta does, they admire Delta and assume Delta knows what it’s doing.

There’s no fundamental reason that — even if American and United were going to following in giving more miles to higher fares — that they’d follow Delta in giving up on average fares. When Delta changed to revenue-based earning they set the ‘break even’ at about 20 cents per mile, significantly higher than their revenue average.

Meanwhilee nearly all United fares still earn 1 mile per mile flown when crediting to Singapore Airlines KrisFlyer. Nearly all (non-Basic Economy) Delta fares earn 1 mile per mile flown when crediting to Czech OK Plus. Nearly all American fares earn 1 mile per mile flown when crediting to Cathay Pacific AsiaMiles or to Etihad Guest.

Though Sorenson’s report carries all of the appropriate caveats about picking a strategy that is best for any given individual program, and offers some detail on Alaska Airlines continuing to use distance-based accrual as a differentiator in its battle with Delta in Seattle, the implication that a less than careful reader might take away is that ‘the debate has been settled’ about how best to rebate customers. Quite the contrary — distance is imperfect, total revenue is imperfect, and the airlines haven’t yet made much progress in the best way to incentivize customer behavior.

Distance was simply easy to measure and used to be more correlated with revenue than it is today. It’s an imperfect proxy for wallet share, how much of your business you give to an airline.

For 30 years airlines have been claiming that the changes they make to frequent flyer programs have been to ‘better reward higher-spending customers. Yet when United, Delta, and American moved to a revenue-based program they made the miles even high fare customers earned worth less by raising the number of miles it takes to redeem for most awards. The revenue-based shift is as much about spending less on marketing, when planes are running full, then incentivizing filling seats. Delta says this change doesn’t drive their revenue premium

Calling total revenue a good measure of customer value is to misunderstand the difference between marginal and average, and to fail to understand opportunity cost. The person who buys full fare tickets but doesn’t pick their own travel provider (either because they ignore it, their corporate travel department buys the tickets) isn’t going to spend more money on an airline because the loyalty program is slanted to high revenue travelers.

Yet a program can influence both choice of travel provider and total amount of travel. A business traveler is less reluctant to travel if she’s being rewarded for it. They may not fight against travel when they aren’t just getting their expenses covered but they have something to cash in later. Without a rewarding scheme, business travelers become more reluctant to travel.

  • A high fare customer isn’t the same thing as a profitable customer

  • You don’t want to “reward” a high fare customer you want to “incentivize” additional business you wouldn’t otherwise get. In other words you want to increase profit (or avoid loss).

It’s interesting that a $300 Los Angeles – Las Vegas fare is rewarded under a revenue-based system the same as a $300 Los Angeles – New York one (and the same as a $300 Los Angeles – Tokyo fare).

  • A flyer may buy one expensive ticket with an airline because they’re the only one that flies non-stop on the route. Does it make sense to reward them, if the customer was going to buy the ticket either way?

  • In general a high revenue passenger is probably better for an airline than a low fare one. But a high fare passenger may trade off with another high fare passenger (for instance they both buy the last seat available on a flight). That high fare customer wouldn’t actually be profitable in an economic sense (opportunity cost basis).

  • A low fare passenger may fill empty seats and be pure profit — or they may ultimately displace a high fare passenger and be very costly if the airline didn’t get their revenue management right.

  • Low fare customers may also engage with an airline’s ancillary products. Base airfare isn’t the only contribution to revenue that matters, and other products are often higher margin than the actual airline seat (priority boarding, seat assignments, extra legroom, checked bags etc). Third party partner customers are profitable too. A member who carries an airline’s credit card and uses it, credits points for their non-air travel to the program, and uses their shopping portal may be a profitable customer.

  • The program needs to try to influence incremental business. An airline may reward a high spend customer but not get additional business from them than you would otherwise have gotten. They might be able to move the needle with some of your other customer segments.

While knowing nothing else about a customer it’s a reasonable guess that the higher revenue customer is more valuable, the whole point of these programs from the beginning was to get better data on customers, to understand and market to them. They were early revolutionary efforts at big data. So ‘knowing nothing else’ isn’t ever a position a loyalty program should be in. And airlines should at least try to do better.

Ultimately if you take your business seriously, instead of just repeating platitudes, then frequency and wallet share matter — and especially driving greater frequency and increased wallet share — and not just total revenue.

About Gary Leff

Gary Leff is one of the foremost experts in the field of miles, points, and frequent business travel - a topic he has covered since 2002. Co-founder of frequent flyer community, emcee of the Freddie Awards, and named one of the "World's Top Travel Experts" by Conde' Nast Traveler (2010-Present) Gary has been a guest on most major news media, profiled in several top print publications, and published broadly on the topic of consumer loyalty. More About Gary »

More articles by Gary Leff »



  1. Wow. That was overlong and, in parts, borderline incoherent. VFTW could really use an editor, or at the very least, a spellchecker.

  2. How about a program that awards a random number of miles for each trip? It’s like a box of chocolates: You never know what you will get. Most customers might enjoy the sweepstakes aspect. It could build more loyalty than, for example, BA’s 25% of actual miles for deeply discounted tickets.

  3. I really think this whole issue is a lot simpler than all that. What has changed is technology and the ability to have more immediate and complete information about what is actually driving revenue and profit. I don’t think it is any mystery that spending is what you want to maximize from your customers. Miles flown were likely only used out of simplicity/necessity in earlier days where technology didn’t track everything else so completely.

  4. What nonsense. Everyone in the 99% are ignored to try to pry more money out of those who have it in the 1% who don’t even need to bother with miles because they’re so filthy rich after 30 years of policies that funneled all the wealth to them.

    Meanwhile the dupes who voted for these policies are strapped into backboards without an inch to move when they fly with no way out because they don’t have the disposable money the airlines think they’re going to squeeze out of a turnip.

    All of this because more than half of them voted for policies that funneled all the wealth to the top – disguised with nationalist, racist, sexist, homophobic appeals, revealing again the open wound that never heals that half of the US are haters who will vote against their own interests to get their hate off.

    BTW after Bush they now teach this in the schools of the modern nations like UK, Spain, Australia, Japan and Germany who have sharply higher quality of life ratings from their people. They even know the name of these dupes who forfeited en masse the American Dream. Do you?


  5. @Gary you often point out the problems and mistakes but you rarely if ever give any kind of solution.

    Tell me again that you’re not a glibertarian.

  6. This post doesn’t make sense. Which is a shame, because I dislike revenue-based earning as much as the next person

  7. @Rob technology isn’t being used here

    @Gadfly I’m not endorsing mileage-based, suggesting both have problems, but that the monkey see monkey do industry isn’t innovating they’re copying and telling themselves they’ve solved something they haven’t. This is not in any way political.

  8. With Delta, You are lucky they don’t anesthetize you on the plane and take one of your kidneys.

  9. it became a blessing in disguise for me. instead of wasting time doing mileage runs domestically, now i’m fully taking advantage of all the cheap international flights. Just booked BCN-OAK this morning for $105 a.i.

  10. I think that the data they are collecting is not telling the true story. It’s taken me a couple of years now to finally pick flights based on ‘value’ rather than chasing the AA experience elite wheel. It’s funny the majority of my flights are on Delta One- because of the price/service and lounge and my loyalty program of choice is Alaska. Experience AA from 2009-2016. Flying a little bit less but spending much less money. Unfortunately also earning much less miles. Also topping up paid flights with award flights and buying Alaska and Avianca miles. It’s a complex formula but bottom line the airlines are getting less $$ from me per mile flown, but I am also getting less free perks and upgrades. I now pay (awards/paid January or F) to fly up front.

  11. Sorry could you replace Experience AA with Executive Platinum on my comment. Phone auto correct problem.

  12. As a frequent east coast shuttle user (high fare) low miles user I am glad the airlines half adopted revenue components as a metric

    Although not perfect , I was essentially subsidizing low fare high mileage award collector

    As a business owner it’s about loyalty and revenue

    Finding the right formula is always the challenge

  13. Between this post, the other posts cited therein, and Sorensen’s 13-page article, there is much to sort through, absorb and respond to. I’ll save a full reading of Sorensen’s article for my next long international flight when there is time and peace and quiet to think.

    Here are a few initial points, observations, and thoughts about loyalty and revenue-based ff programs.

    (1) FF programs, whether spend or mileage based, have no correlation to loyalty. Being loyal to an airline isn’t the same as being loyal to one’s spouse. As a DL Diamond and AA EXP simultaneously, I’m clearly not loyal, in the traditional sense, to either program yet these airlines claim that I am.

    (2) Related to the first point, I’d think that the goal of a ff program is not only to keep your customers but to attract customers from your competition. There is a benefit not only to the revenue earned from a passenger but from the denial of revenue to the competitor.

    (3) Sorensen’s statement on page 7 of his article that “Every airline, except American, Delta, and United, currently uses the type of fare purchased by the consumer to help determine the final tally of a member’s mileage or point total” is absolutely false. The type of fare – economy, full fare economy, or discounted or full fare business and first class – most certainly factors into a member’s mileage and point total on the US Big 3.

    (4) The value or profitability of a passenger is a terribly complex determination. Most would say big-spending business passengers who cost an airline more to serve are way more profitable than lower fare economy passengers who cost much less to serve and are the major source of highly profitable ancillary fees, services and revenue streams. Revenue is important, as Sorensen states, but one cannot overlook costs. I’d argue that any flight with a full premium cabin is unprofitable without the folks in the back. Also while interest rates are now extremely low, how much of a benefit it is for airlines to get payment months in advance from lower-fare customers? In addition to the free use of someone else’s money, those low fares may wind up being changed for a high cost or not used at all.

    (5) The market may be the best way to determine which ff accrual method is better. But that would require competition between the two models, and since all of the Big 3 ar revenue based, that’s not going to happen. On page 9 of his article, Sorensen delves briefly into the battle between DL and AS in Seattle. Given DLs numerous advantages, that’s not a fair fight and not really a good market test. The fact that AS seems to be holding its own against big bad DL that will pull any trick in the book to crush its former partner might be evidence that AS’s ff program is working in its favor.

    (6) The Big 3 do not reward the biggest spenders the most. They cap accruals at 75,000 miles/ticket regardless of how much the customer spends. Really big spenders don’t get rewarded more than plain old big spenders.

    (7) As I’ve said a few times before, revenue-based programs create an enormous employee/employer conflict of interest. For example, a normal conflict is where an employee receives some personal benefit by directing company business to a certain supplier of goods or services that the company needs (hopefully). With revenue-based programs the employee not only receives a benefit by using a Big 3 airline but that personal employee benefit is larger the more the employer pays for the ticket! Remarkable.

  14. Ignore the haters, Gary. Your valuable breakdown on Jay’s report is lost on most folks.

    While I don’t agree revenue based programs are the king (in fact they are far from it), it’s important to consider who benefits from the exchange. In the case of the FFP, it earns more from the host airline in mileage sales (so a win – and certainly checks the incremental revenue box), the airline will also pay less on lower yield tickets – so the overall net result of a plane full of pax would be lower overall costs to the FFP as planes are coach/economy class heavy.

    Does a program engineered this way engender additional loyalty and create new revenue that otherwise would never exist? This is where propensity data modeling and pax scoring (you allude to this in the article as something airlines should do – and they most certainly DO already, without pax insight) comes into play, and is not public information which can be critiqued by outside eyeballs.

    Ultimately – revenue based programs cut out traditional mileage runs and create a more honest view of customer spend. In this sense, the business logic is sound. The share of wallet, increased engagement frequency, and incremental revenue questions are all addressed by frequent flyer programs through other means.

    From an outside perspective, revenue programs look bad. Internally, they paint a different story, however, fundamentally I agree with your sentiment that there are better solutions available. The problem is, that, once a program releases it, the others will copy.

Comments are closed.