Disruption Is the New Normal. The Real Failure Is What Happens Next.

Delays, cancellations, and diversions have exposed a fragmentation problem that runs

from the first decision to the last payment — and the airlines that unify that chain will

win the passenger.

Part 1 of a two-part series on IROPS.

A departures board at the San Francisco International Airport on 7 November 2025. Photograph: John G Mabanglo/EPA

Ask any traveler heading to an airport this summer what they expect, and the honest answer

is no longer “a smooth trip.” It is “a delay, probably.” A May 2026 Hopper Technology

Solutions survey of more than a thousand U.S. travelers, reported by Fortune, found nearly

nine in ten travelers planning to fly in the next year worried about delays or cancellations —

with roughly one in four describing themselves as extremely concerned. Passengers are

building buffer days into itineraries, paying extra for flexibility, and booking around the

assumption that something will go wrong.

In other words, passengers have already accepted what much of the industry still treats as

an exception: irregular operations are no longer irregular. By Hopper’s count, the

number of “significant disruption days” in the U.S. has roughly doubled since before the

pandemic, and the seats affected each year have climbed from roughly 50 million in 2019 to

58 million in 2025. Geopolitical airspace closures, fuel volatility, air traffic control

constraints, and a network running at or above capacity with almost no slack have made

large-scale disruption a structural feature of the system, not a bad-luck event.

That leads to a conclusion most airlines have not fully internalized: when disruption is

the baseline expectation, recovery becomes the product. And disruption is far

more than the cancellation that makes the evening news. It is the three-hour delay that

breaks a connection, the rolling delay that strands a family overnight, the diversion that

puts an aircraft — and two hundred passengers — in the wrong city. On a single heavy day

this summer — June 26, 2026 — U.S. hubs logged roughly 2,600 delays against fewer than

100 outright cancellations. Delays are the everyday grind of disruption; cancellations are

just its loudest form. The airline brand is no longer defined by the disruption itself —

passengers largely understand that weather and airspace are outside anyone’s control. It is

defined by what happens in the hours after the disruption begins. And that is exactly where

the industry is weakest.

The Recovery Is Automated. It Is Also Fragmented.

The specialist ecosystem that has grown up around passenger recovery is real, and much of

it works. Digital voucher platforms have genuinely retired the paper voucher at many

carriers. Disruption-payment specialists have put money on a passenger’s phone in minutes.

Re-accommodation platforms have pulled the seat, the hotel, and the meal behind a single

front end. Each of these is real progress — and each covers its slice of the recovery and no

more. Look closely at almost any of them and the same architecture appears underneath:

specialist third parties, assembled behind one interface, with the seams managed rather

than eliminated.

Naming that architecture is the key to understanding this market, because two models are

now competing in it: integration versus unification. Integration assembles the recovery

from best-of-breed pieces — the voucher from one provider, the hotel through another, the

settlement on a third rail — and hides the joins behind a single screen. It is a pragmatic

model, often the right first step, and it is where almost the entire market sits today. But the

seams are still there, in the settlement lags, the data handoffs, the timing gaps between the

voucher that arrives and the hotel that doesn’t. Unification runs the chain natively: one

platform detects the event, decides the entitlement, issues the payment, books the room,

and settles with the supplier — no seams to hide, because there are no handoffs to make.

And it is no accident that the platforms attempting genuine unification are built payments-

first. Payments are the connective tissue of the recovery chain — the one element that

touches every step, every party, and every seam. A voucher is a payment. A hotel room is a

payment. Crew accommodation, ground transport, the refund, the interline settlement —

payments, all of them, each currently running on its own rail at its own speed. Modern

payment technology is what makes unification executable rather than aspirational: wallet-

native disbursement puts the entitlement on the passenger’s phone the moment the event

fires; virtual card rails let the platform source and settle a hotel room in one motion, paying

the supplier at the moment of service instead of weeks after it; real-time settlement closes

the loop with suppliers and partner carriers while the disruption is still live. Solve the

money movement and the seams lose their power — because most of what leaks through a

seam is a payment. That is why the recovery stack of the future is being assembled on

payment rails, not bolted onto them.

And unification is no longer a hypothesis. In our client work we have evaluated a

platform, live in the market today, that has automated everything except the seat. One

system detects the event and decides the entitlement — configurable against the full

regulatory map: EU261, the DOT rules, baggage obligations and beyond — and then

executes it end to end: meal, hotel, and ground transport sourced, booked, paid, and settled

natively, with the funds arriving wallet-native on the passenger’s phone. It runs the crew’s

care — their hotels, meals, and ground — on the same spine as the passenger’s. And because

it sources and settles accommodation and transport itself, it can do what the integration

model structurally cannot: remove the accommodation and ground brokers from the chain

entirely — and with them, their commissions, their batch settlement, and their seams. We

are deliberately not naming it, because the point of this piece is the architecture, not the

vendor. But notice the shape of the market this reveals. On one side sit the re-

accommodation leaders — genuinely strong platforms that have solved the passenger’s seat,

and solved it well — but at premium cost, without the crew, without a line upstream into the

fleet and operations decisions, and without the care and payments chain downstream. On

the other side sits a platform that has solved everything but the seat. The two halves of full

unification exist today — separately, each incomplete without the other, and neither

reaching the decision itself. The question in front of the industry is no longer whether the

model is possible; it is how fast the halves come together and become the expected standard.

The passenger, meanwhile, experiences the difference every time.

How does an airline tell which model it is actually buying? Not from the brochure — every

vendor’s diagram looks unified. It shows up in the answers to five questions. Does the

voucher load to the passenger’s mobile wallet, or only to a QR code and a closed merchant

list? When the recovery includes a hotel, is the room sourced and settled natively, or

fulfilled by a subcontractor the airline never sees? Does a single trigger fire the payment, the

hotel, and the ground transport simultaneously — or in sequence, with the passenger last?

Who settles with the hotel, and how long after the stay? And does the coverage hold outside

the vendor’s home region, on the airline’s worst night, in a currency the platform didn’t

grow up in? Wherever the answer is “it depends” or “through our partner,” that is a seam —

and the passenger will find it before the procurement team does.

And beneath the whole passenger-facing layer sits a supply chain with problems of its own.

In our conversations across the industry, hoteliers are blunt about disrupted-passenger

business: it arrives last-minute, consumes the last rooms — precisely the highest-yielding

inventory they would otherwise sell at peak rates — and then pays a commission to the

accommodation intermediaries who broker the booking, while settlement lags behind the

stay. The hotel is squeezed three ways: yield, commission, and payment timing. There are,

frankly, too many hands in the pot — and yet every hand has been necessary, until now,

because sourcing hundreds of rooms across a city at eleven o’clock at night during a mass

disruption is not a capability any airline possesses in-house. As we noted above, this is

precisely what native sourcing and settlement changes: the broker layer becomes

removable, the hotel gets paid at the moment of service instead of weeks after it, and the

too-many-hands paradox is revealed for what it is — not a law of this market, but an artifact

of the integration model. Until that model is replaced, the paradox stands, and it puts the

very room supply that airlines depend on for their worst nights at quiet, structural risk.

And notice what runs through every seam in this landscape. Almost every step in the

passenger’s disruption journey ends in a payment:

• A refund for the cancelled — or significantly delayed — flight.

• A meal voucher — pushed digitally to the passenger’s phone at most carriers now, with

the gate agent as the human fallback for the passenger without a smartphone, without

connectivity, or without the app.

• A hotel room and ground transport for an overnight stranding.

• Reimbursement for out-of-pocket expenses the passenger fronted.

• Settlement with another carrier when re-accommodation crosses airline lines.

• And the airline’s own people: hotels, meals, and ground transport for the displaced

crew — a disrupted party the public never thinks about, with a payment and logistics

stream of its own.

The passenger never sees the operations control center. They never see the crew re-

optimization engine. What they see — what they remember, and what they post about — is

how long it took to get a room, a meal, and their money back. The disruption is operational.

The experience of the disruption is financial.

IROPS Is a Payments Problem in Disguise

This is the part of the IROPS conversation almost nobody in the industry is having, and it is

the part that now carries real regulatory teeth.

In the United States, the Department of Transportation’s automatic refund rule is fully in

effect — and critically, it is triggered by delays, not just cancellations. A domestic delay of

three hours or more, or an international delay of six, puts the airline on the clock just as

surely as a cancellation does. When the passenger declines rebooking, the refund must be

automatic — no forms, no phone queues — issued in the original form of payment within

seven business days for card purchases, and airlines can no longer substitute vouchers or

travel credits unless the passenger affirmatively chooses them. Non-compliance carries fines

that can reach tens of thousands of dollars per violation. And Washington, in fact, came

within a signature of EU-style tiered cash compensation — $200 to $775 per passenger for

airline-caused disruptions — before the proposal was withdrawn in late 2025. The mandate

is gone; the political appetite that produced it has not, with legislation to reinstate it already

introduced in the Senate. The regulatory pendulum swings — and airlines that build

recovery capability only when compelled will be caught flat-footed when it swings back.

Meanwhile, compensation obligations already bind U.S. carriers today on parts of their

international networks: EU261 on their flights departing European airports, and Canada’s

Air Passenger Protection Regulations on flights to and from Canada.

In Europe, EU261 has been the benchmark for two decades — and after thirteen years of

legislative deadlock, the European Parliament and Council reached political agreement in

June 2026 on the first major reform of the regulation: standardizing claims procedures,

codifying the list of extraordinary circumstances into law, and requiring airlines to

proactively inform disrupted passengers of the cause and their rights within 96 hours. The

new rules, once formally adopted, are not expected to take effect before late 2027; until

then, the existing regime — with its €250–€600 compensation — remains fully in force.

Enforcement bodies handled record complaint volumes over the past two years, and airlines

are settling valid claims faster to stay ahead of regulators.

Read those developments together and the message is unmistakable: regulators have

turned payment speed into a compliance obligation. The option to be slow with the

money is gone.

Yet the plumbing underneath has not kept pace with that obligation. Refunds still crawl

through settlement processes that were never connected to the operations systems that

trigger them, and duty-of-care money — hotels, crew accommodation, ground transport —

moves through separate platforms on separate rails, compounding the hotel squeeze

described earlier. Every one of those seams converts an operational event into a brand event

— for the worse.

Caught in the Middle: The Cost-Versus-Brand Tightrope

Here is the position airlines actually find themselves in, and it deserves more honesty than it

usually gets. Some disruption is genuinely outside the airline’s control — weather, air traffic

constraints, airspace closures, fuel shocks — yet the passenger blames the airline anyway,

and the U.S. refund obligation applies regardless of cause. But some disruption is the

airline’s own: mechanical issues, crew availability, operational failures — events that fall

squarely outside any act-of-god provision, and where the exposure is heaviest, since duty-of-

care commitments and the proposed compensation regimes are aimed precisely at airline-

caused disruption. And within that “controllable” category sits an uncomfortable truth the

public rarely appreciates: many of those disruptions are the direct product of the airline’s

highest obligation — safety first. The aircraft held at the gate over a warning light, the crew

timed out rather than pushed past duty limits — these are airlines choosing to disrupt rather

than compromise safety. Yet the system treats that decision exactly as it treats poor

planning. The airline is squeezed from every direction at once: blamed for what it did not

cause, liable for what it did, and generating disruption through the one discipline it can

never trade away.

And once the disruption hits, every recovery becomes a live spending decision made under

pressure. How generous is the meal allowance? Does this passenger get a hotel, or a seat in

the terminal? Do we rebook on a competitor — and pay for it — or hold the passenger for

our own next flight tomorrow? Multiply those decisions by thousands of passengers on a

bad day, and the tightrope comes into focus: how much am I willing to spend on care,

versus how much am I willing to risk the brand by saving money in front of a

customer who is already having a terrible day?

Because that is the context that makes disruption handling so unforgiving: the passenger is

already upset before the airline does anything. The disruption itself set the emotional

temperature. From that moment, the airline’s handling does one of two things — it redeems

the situation, or it compounds it. An airline that manages the recovery badly is not

remembered as the victim of a thunderstorm; it is remembered as the villain of the

aftermath. The fifty dollars saved by squeezing the care budget can cost a customer worth

thousands in lifetime value — and in the social media era, the story of the bad night travels a

lot further than the passenger who lived it.

And there is a second human being on that tightrope, wearing the airline’s uniform. Every

seam the passenger experiences, an employee absorbs face-to-face. The gate agent did not

cancel the flight, did not size the voucher, cannot see why the rebooking engine made the

choice it made, and cannot say which vendors will accept the QR code on the passenger’s

phone — yet they stand in front of two hundred angry people as the airline’s only visible

representative, improvising answers a fragmented system cannot give them. The costs are

real and compounding: burnout and attrition concentrated in exactly the roles airlines

struggle hardest to staff, service quality collapsing at the precise moment the brand is most

exposed. The gate during a mass disruption is where frustration finds a human target.

Automation done properly does not replace the agent — it removes the worst hour of their

job. When the entitlement, the payment, the hotel, and the answer arrive on the passenger’s

phone before they reach the podium, the agent is freed for the passengers who genuinely

need a human — the unaccompanied minor, the traveler without a smartphone, the family

no algorithm anticipated. For airlines fighting to hold onto frontline talent, a unified

recovery may be the ROI argument the CFO hears first.

The uncomfortable truth is that fragmentation makes this tightrope far worse than it needs

to be. When the voucher platform, the refund process, the hotel program, and the rebooking

engine all live in separate systems, the airline cannot see — let alone manage — the tradeoff

in real time. Cost control happens blind, and the passenger experience happens by accident.

A unified recovery dissolves much of the dilemma: done properly, generosity gets cheaper

and the experience gets better — at the same time. The cost-versus-experience tradeoff is

largely a symptom of the fragmented stack, not a law of nature.

A Case in Point: American’s $12 Answer

If you want to watch the tightrope being walked in public, look at what American Airlines

announced in late June 2026, on the heels of one of its roughest disruption stretches of the

year. American extended automatic meal vouchers — previously an elite perk — to every

AAdvantage member, and updated its systems so that an entire flight’s worth of vouchers

triggers at once the moment a qualifying delay hits, delivered digitally to the passenger’s

phone. Credit where it is due: an operational event fires a payment automatically, with no

queue and no phone call, at one of the world’s largest carriers. To be clear about what that

proves: not that the architecture can be built — as we noted above, unified recovery already

exists in the market. What American proved is something different, and in its own way just

as important: that a major legacy carrier can wire its own operations systems to automatic

payments at fleet scale. The technical barrier fell elsewhere; the institutional barrier just fell

here.

Now look at what was actually delivered through that plumbing. The voucher is twelve

dollars — an amount that, at 2026 airport prices, struggles to clear a fast-food counter;

passengers were quick to report terminals where no hot meal came in under it. The care is

gated behind joining the loyalty program — free, yes, but it quietly converts a duty-of-care

moment into a membership-enrollment and data-capture event, while the infrequent flyer

— the very passenger whose single bad experience will define the brand forever — gets

nothing automatically. And the redemption experience betrays the seams beneath the

automation: vouchers arriving after boarding and unusable for hours, accepted only at

certain vendors that even the airline’s own staff could not identify. Nor does the trigger

reach as far as the announcement implied: the voucher fires only for controllable delays —

three hours or more, with at least ninety minutes of it the airline’s own doing. The weather-

stranded passenger, the most common kind, gets no automatic voucher at all. Hotels,

meanwhile, remain largely agent-mediated and excluded entirely for weather events, and

the refund is a separate process on a separate system.

And there is one more layer here — one we will offer plainly as an informed read from years

inside loyalty economics, not as a reported fact, because the commercial plumbing is not

visible from outside. This design has the unmistakable shape of a co-brand structure. Care

gated behind program membership grows the enrolled base — every disrupted passenger

becomes a loyalty acquisition. A closed list of participating vendors is what negotiated

funding looks like. And it would surprise no one who has sat inside these deals if the twelve

dollars is not really American’s money at all — absorbed instead by the co-brand

partnership or the participating vendors themselves, in exchange for guaranteed foot traffic

and spend routed into the program’s ecosystem. If that read is right, the “care” is not a cost

center. It is a monetized loyalty product wearing a duty-of-care costume. As commercial

engineering, it is genuinely clever. As recovery, it is not recovery.

So here is the honest scorecard. The most visible recovery-automation move by a major U.S.

carrier in 2026 is a twelve-dollar fast-food voucher with a membership requirement. The

issuance was automated; the experience was not. One payment type, in one channel, hedged with conditions — a fragment, doing exactly what fragments do. You can even read the org chart in the product: operations built the trigger, finance sized the amount, loyalty added the gate — and, if our read is right, loyalty may have funded the whole thing. Three departments, three agendas, one voucher. American demonstrated that a legacy carrier can make the connection — and demonstrated, in the same stroke, how far a fragment falls short of a recovery.

The Economics Make This Urgent

The financial stakes are stark. By IATA’s own industry outlook, the average net profit is in

the range of single-digit euros per passenger. A single major hub disruption can erase weeks

of operating profit. IROPS has migrated from the operations budget to the balance sheet.

Meanwhile, watch who is monetizing the gap airlines have left open. Travel fintech players

have built highly profitable businesses selling disruption protection, flexibility products, and

instant rebooking — Hopper reports that fintech products now generate 45 percent of its

total profit, per Fortune’s May 2026 coverage. Third parties are earning margin, and

passenger loyalty, on the airline’s worst day. If the airline does not own the moment the

money moves, someone else will — and they will take the customer relationship with it.

The Holy Grail of IROPS: True End-to-End

Everything to this point has described the passenger-facing half of the problem. But the true

end-to-end — what we would call the Holy Grail of IROPS — is far bigger than the

passenger. It begins before the passenger knows anything is wrong, at the decision itself —

cancel, or delay? — and it runs across eight domains that today live in eight different

systems: fleet, crew, operations, ground operations, flight operations (gate management,

slot management, routing, and more), re-accommodation, passengers, and recovery.

Its defining capability can be stated in one line: it decides predictively and executes

simultaneously. It weighs both paths — delay versus cancel — knowing the full

downstream price of each before the decision drops, and then fires everything that follows

at once, passengers and crew together, decision to delivery. Recall the tightrope: the cost-

versus-brand tradeoff is agonizing today because it is walked blind. The Holy Grail turns it

into a calculated trade, made once, with full information — instead of a thousand

improvised judgments under fire at the gate.

How the industry’s systems grew this fragmented in the first place, why the eight domains

still cannot speak to each other, and where the opportunities lie to unify them — that is a

subject that deserves, and will get, a full article of its own. Part 2 of this series takes

that journey. Stay tuned.

The Strategic Imperative

Disruption is not going away. Passengers know it, regulators know it, and the operating

environment guarantees it. The competition of the next decade will not be won by the airline

with the fewest delays — no carrier controls the weather, the airspace, or the fuel market. It

will be won by the airline that anticipates, plans intelligently, reacts proactively — and

ultimately recovers best. And recovery runs the full length of the cascade: from the quality of

the first decision to the speed of the last payment.

The carriers that pursue a united, end-to-end recovery — rather than accumulating one

more point solution — will turn their worst operational days into their strongest loyalty

moments. The rest will keep paying for disruption twice: once in the operation, and again in

the customer they lose at the seams.

The disruption is inevitable. The fragmented recovery is a choice. Contact The

Cileo Group to discuss how to navigate the IROPS vendor landscape and build

toward a truly end-to-end recovery — before your passengers, and your

regulators, decide for you.

By: David Palmieri