I remember reading about how the major airlines now are more of a "bank that happens to have planes," due to the loyalty programs being worth significantly more than the airline. Delta Air Lines earned $8.2 billion from American Express in 2025, surpassing ticket sales revenue. [1]
I primarily use my favorite's airlines credit card because it gives me perks such as priority seating, and free checked bags. I am pretty certain that the credit card fees (that is passed on to the merchant) does not come close to the value that I gain for my credit card loyalty. It is a stupid game that I am forced to play, because the credit cards also provide other benefits, such as fraud protection.
I am wondering right now if "Spirit Air 2.0" even has a fighting chance if they are not able to subsidize operating costs by also being a credit card company.
>Delta Air Lines earned $8.2 billion from American Express in 2025, surpassing ticket sales revenue.
Just to be clear, that isn't what the article says. It says more than what "most" airlines generate in ticket sales. Not Delta, or any major US carrier. As interesting as that sounds, it couldn't logically make sense and it only represents about 15% of Delta's revenue. It's not even a straightforward revenue stream, it works for profitability because they are able to book most of the revenue immediately and able to mark down the future expense because of how loyalty rewards are obligated.
Yes, but how does stating the obvious - that airlines make almost all their money from flying planes, and that different lenses of the loyalty program's intrinsic value is accounting parlor tricks, and that the main reason they like the credit cards is because it drives people to fly with their loyal choice more - how does that drive more listeners to your podcast? Checkmate, reasonable guy.
This isn't really a bad thing. Any company that monetizes credit cards can only do so because of their real, core product. They aren't really just banks like people claim. If they didn't fly people places reliably the whole thing collapses.
It's really just a surprising morph of their economic model in the post regulation era.
Many studies show that consumers spend more when using credit cards, compared to both cash and debit cards, and it's not obvious to me that the 2% will win out.
In the case of air travel where you're, presumably, going to travel either way, I think it can work out. But it only works out if you travel A LOT. Even a moderate amount of travel means you don't win out with the cost of membership.
That's one free round trip international flight per year in a lot of cases. Plus sometimes other benefits like theft insurance, warranty extension, phone insurance, etc.
You're subsidizing everyone else if you're not trying to get the best loyalty program.
Welcome to capitalism. Do you refuse to buy items on sale at the grocery store because they are subsidized by the more premium items? Do you refuse to buy the rotisserie chicken because it’s a loss-leader priced to get you into the door?
You’re taking a 2%+ loss on every purchase if you’re not playing the game. But you better carry debit cards and cash because some vendors charge extra for credit cards, and some charge extra for using cards at all!
This sort of reminds me of a gas station that only sold gas at the pump. No convenience store, no smokes, no snacks, no cokes. I think it was Swifty? I know it was yellow signage.
Anyway, point is they failed and went under and my recollection is that just selling gasoline alone was not profitable. The extra coin comes from selling snacks, beer, smokes, etc.
I used to work for a company with about 1200 gas station/convenience stores. They tracked "Light product break even". Basically the profit to earn on fuel to make the store break even. It was like 2 cents per gallon most of the time. So they could be super competitive on fuel and still be profitable. It was pretty crazy to see that.
And it was Swifty - definitely a no frills experiences!
This loyalty program is the business is oversold imo, done to death by every content creator. It's the data, the data blah blah
The $8.2billion from American express pays basically is buying tickets and ticket extra, it buys them some points, lets ignore multiples for now, it buys them 8.2billion points, which they give to customers which then buys tickets.
If Spirit accepts USDC instead it wouldn't be that much different.
> I am pretty certain that the credit card fees (that is passed on to the merchant) does not come close to the value that I gain for my credit card loyalty.
Generally it's the interchange fees that fund reward programs (charged between banks), not the merchant fee.
It generally depends on the contract the merchant has with payment provider:
- some have relatively high merchant fees to cover for interchange fees
- others (generally called IC+) have the merchant pay the IC fee plus some other (generally much smaller) fee to the payment provider
In both cases it's the merchant that ends up paying them. It's not a concidence that in Europe (where there are caps to IC fees) the fees that merchants pay are generally lower.
this isn't unique to airlines. this applies to all exceptionally mature companies/industries who believe there is no more room for growth. any significant profit they make gets paid out as dividends to investors who then put the money elsewhere instead of reinvesting into the company
This looks like yet another basic key value store.
Benchmarking is a complicated problem, but FoundationDB claims 300,000 reads per second on a single core. TigerBeetle claims 100k-500k TPS on... Some kind of hardware?
The design with hot/cold storage makes it much more interesting than FDB for some use cases. FDB is an excellent DB with very strong operational guarantees, TigerBeetle seems to be specialized for financial data and to optimize perf/cost ratio.
This is a nice article, and I appreciate the examples. The next problem to solve is how to persist state on disk across two different accounts after a transfer has been done.
microvms as espoused by things like firecracker offer full machines but have tradeoffs like no gpu (which makes it boot faster)
hyperlight shaves way more off - (eg: no access to various devices that you'd find via qemu or firecracker) it does make use of virtualization but it doesn't try to have a full blown machine so it's better for things like embedding simple functions - I actually think it's an interesting concept but it is very different than what firecracker is doing
FoundationDB has been growing as my favorite database lately. Even though it is only key-value store.
Out of curiosity: what are the scale limits of FoundationDB? What kind of issues would it start to have? For example, being able to store all of Discord messages on it?
I see blog posts of Discord moving to Scylla and ElasticSearch, but I wonder if there would be any difficulties here.
There are a lot of strict limits so AFAIK everyone uses FoundationDB for fast, consistent, highly-available metadata while doing replication/storage of actual data elsewhere (such as in S3).
In their data modelling pages they mention you should break up rows into separate keys per column. Or separate keys per field in a document. This is indeed how many databases model rows on a distributed kv store. So this might be how they achieved 100TB.
However you still have the issue of any single key-value needing to be in their limit. (But it's not like people typically store enormous blobs in Postgres or MySQL either I think?)
The documentation is woefully out of date, sadly. Despite the code being in active development no one is touching the public docs. Though I don’t know for sure, that limitation was probably written something like 10 years ago.
Before you execute the query, you should be able to query any of the data, and after you execute the query, none of the data should be available. The mechanisms to make a transactional database is tricky.
Some databases, like CockroachDB, provides some built-in TTL capabilities.
But also- if you are having to delete huge ranges of data and do not care about consistency, you are probably looking at an analytical workload, and there would be better databases suited for that, like Clickhouse.
As written, that's not required. The data should not be retrieveable by query, but ACID only specifies what happens at the client-server boundary, not what happens at the server-storage boundary (Durability prescribes that the server must persist the data, but not how to persist it). A database that implements DELETEs by only tombstoning the row and doesn't discard the data until the next re-index or vacuum operation would still be ACID-compliant.
As usual, there is a spectrum of data safety vs. performance.
Redis is at the "very fast, but unsafe" side of the scale.
ScyllaDB for me is in the middle of being high performance key-value store, but not really supporting transactions. FoundationDB is another one that I would consider.
Depends on the kind of safety you’re looking for. Redis is entirely safe from concurrency issues because it’s single-threaded. It supports an append-only file for persistence to disk.
I primarily use my favorite's airlines credit card because it gives me perks such as priority seating, and free checked bags. I am pretty certain that the credit card fees (that is passed on to the merchant) does not come close to the value that I gain for my credit card loyalty. It is a stupid game that I am forced to play, because the credit cards also provide other benefits, such as fraud protection.
I am wondering right now if "Spirit Air 2.0" even has a fighting chance if they are not able to subsidize operating costs by also being a credit card company.
[1] https://www.thestreet.com/personal-finance/delta-air-lines-m...
reply