The Lev

When looking at passenger profitability (as opposed to marginal profitability) you can't just look at incremental costs.

canadiancow

But The Lev 's point was that while there exists a CASM, the CASM on a 77W YVR-YYZ is not the same as a 320 YVR-YYZ, or a BEH YHZ-YQM, or a 789 YYZ-HKG.

If they do it naively, people like FTers will game it. But the smarter they get, the more they reward people who deserve it.

acse17

Please and thank you.

Laughable that I have worry about mileage runs when I regularly double the spend required.

theBeachBoy

So you don't fly 67k miles (1.5x in J) while spending $40k?

acse17

I don't fly paid J except for a few specific infrequent routes. The other sad miserable part of my existence is that our corp travel agent shows me flex fares, accepts the money for the flex fare, and yet over the last two years I've discovered that I have a significant amount of fares that have been ticketed as standard. This year i'm missing over 30k that would have put me over AQM a while ago. Our category manager went down the rabbit hole of trying to get it fixed and we got crickets from both sides. Oh well... at least next year I'm greenlit to fly all comfort.

Sean Peever

How are there 65 posts on a program that didn't change at all ?!?!?!?!?!?!?!?!?!?!?!?!?!

well ok, 75k now gets 12 months wifi option, sorry! 1 change, 65 posts, what ?!?!?!?!?!?!

Adam Smith

Big data and cheap computational power, sure. But with the morass that is AC's IT operation, I don't see it being realistically possible. Think about what that would entail.

Here are some challenges on the revenue side, beyond just using AQD:

• There are some big chunks of revenue that aren't accounted for in AQD - change fees, seat selection fees, checked baggage fees. Maybe other fees that I haven't thought of. Oh, YQ on reward tickets, there's another one. SDC fees. Someone who has $20K of AQD and spends $2K a year on change fees would likely be more profitable than someone who spends $21K of AQD, for instance. And an E35K who shells out for a preferred seat on most trips could be adding pretty meaningfully to their revenue. Could AC track these things? Sure. But the fact that they currently don't include them in AQD suggests there's some challenge in that. So, once again, we're probably trying to integrate different systems and there are likely to be issues.
• If you fly OAL metal on OAL stock, you get no AQD. AC gets no revenue for those tickets (other than a few bucks for selling some Aeroplan miles to the OAL), but how do you assess the contribution to that person's profitability? Especially if the itinerary includes some flights on AC metal, e.g. a UA ticket that has an AC flight then a UA flight? Because people frequently choose *A partners because of connection to AC. Once again, is it possible to come up with some sort of rough approximation for this? Sure. But AC would have to invest a bunch of time in devising the system, then figure out a way to implement it.
• AC pays commission to TAs on some tickets. Someone with $10K of AQD on tickets bought from the AC website has generated more net revenue than someone who always buys tickets through TAs (which is still the case for a lot of business travel). If you're assessing profitability, you should probably be looking at this. Again, the information already exists somewhere at AC, but combining it with the other parameters here will undoubtedly be work.
• Now that AC owns Aeroplan, do you include whatever revenue you generate from people having Aeroplan-linked credit cards?

On the cost side:

• Someone at AC in finance or revenue management would need to calculate some sort of reasonable estimate of cost per route in each CoS. They're already doing that at some level to help them figure out how to price the tickets, so this isn't the hard part, but the database of costs has to be linked to the FFP database so that it can compare your revenue to your cost - making multiple systems talk to each other is an area where we know AC struggles.
• What's the value of an eUp? Do you come up with a system-wide cost that an eUp is worth $12.73 or whatever and then just calculate the cost based on the number of eUps you hand out? Or do you do a route-by-route estimate of what it costs to eUp a passenger and then look at what people have actually used their eUps for? The former is easy but likely to be so generic as to be meaningless. The latter is a better estimate of the true cost, but a lot of work to calculate and then tie together with everything else.
• It's certainly possible to come up a cost estimate for lounge visits based on the average cost, but I question how meaningful that will be, with alcohol being the biggest component. Most of my lounge visits are short, like the cow. If you assign a $25 cost to every lounge visit, I'm going to look very expensive, even though my average visit probably costs less than $1. Meanwhile, some of the rig pigs who show up hours before their YYC/YEG-YYZ redeyes and consume copious amounts of alcohol won't look very expensive because they don't hit the lounge very often, even though when they do, they rack up huge costs. I think the average is the only way to go, but it's going to be quite misleading, and the dollars will be meaningful in many cases.
• What's the cost of Aeroplan miles? Of IKK? Hugely dependent on how the miles are used. Again, do you use an average, or try to figure out something closer to the true cost?

These are just the issues I've thought about in the past 15 minutes or so; I'm sure there are even more that I haven't gotten to.

Could human effort and computing power potentially solve a number of the issues above? Likely. But the amount of effort and dollars are likely to be so significant as to make it not worth the hassle. Either that or it's going to be such a rough approximation that it's not truly calculating that person's profitability, only categorizing them in to one of a few buckets (e.g. highly profitable, moderately profitable, break-even, moderately unprofitable, very unprofitable). That may still be a worthwhile exercise, but I don't think you can really say you're calculating that person's profitability when you're only assigning them a broad category.

I didn't suggest that AC should actually use cost in determining status. But if they wanted to, there's nothing requiring them to publish anything anywhere. Look at UA GS and AA ConciergeKey - no published requirements there at all. In theory, there's nothing stopping an airline from ceasing to publish any requirements and just saying that status in the loyalty program will be awarded by the airline will be awarded purely at its discretion.

I'm not saying that's likely, because that would undermine the purpose of the program, but it's important to remember that FFPs are not an enduring right.

Sean Peever

It's called an ERP. Most businesses have one in some shape or form. Most businesses don't use it correctly.

RangerNS

Ok, a fat guy with heavy free bags might be $10 in go-juice from YYZ to HKG, hugely dwarfed by his share of the weight of the aircraft.

And they both often don't have enough catering, be it "included" J or actual increased spend Y meals. And equally often, throw food out. Throw out the end of wine bottles.

And not even broken down in what they report to the street by class. And infrequently by route. Its an average, and mostly fixed overhead.

Stranger

Sure but then, these programs are by invitation, and it's at the very high end. They still need a "regular" FF program for marketing purposes.

What about the MM program though? They did promise something to last forever, didn't they? :-)

canadiancow

Doesn't the first point there solve the second point? If route XXX-YYY costs $A in Y and $B in J (based on RM having that estimate), and you use Z eUps to go from Y to J, then the eUps were worth (B-A)/Z each.

RangerNS

They promised status by name.

Which could be as useful as German titles of nobility in 1920.

The Lev

Most ERPs are grossly inadequate to do any form of activity based costing.

Given that AC's RESIII system is older than most people here it certainly wouldn't have been able to do it. Amadeus - maybe???

Adam Smith

That's one way to look at it, I suppose. But what about opportunity cost? RM will try hard not to let people eUp if they can sell J, but certainly there are some eUps that do result in J tickets not being sold. (Same concept applies to Aeroplan in general and IKK in particular). Or LMUs or bid-ups not being sold.

canadiancow

So you'd want something like assigning an additional cost of X, where X is the difference between full J and the B value, if the flight is J0?

Could work.

But then I've seen other issues where I wanted to buy 2 J at the last minute, the flight was J1, and there had been a dozen cleared upgrades. And there was another time I wanted 6 J, but it was J4. So it's hard to even know when you've lost a sale based on actual loads.

Just hold all upgrades until the gate. That would solve all the problems.