Episode 007 — "Who Owns the Offer? Sabre Cries Monopoly as the Agents Show Up"
Airlines spent a decade fighting to own their offer. This week three stories mark the same battle line from three angles — and together they reveal who's standing between the airline and the buyer right now.
This week on The V1 Airline Retailing Report, Eric and Steph build one argument across three floors of the same building: the PSS incumbency layer, the MCP agent interface forming above it, and the IATA program running through both. The question threading all three is the same: after NDC and Offer & Order, who owns the offer next?
Sabre's legal complaint against Amadeus has a new frame — and it's more important than the lawsuit. Sabre CEO Kurt Ekert accused Amadeus of using Altéa PSS control to block competing Offer and Order solutions, and announced regulatory and legal action. The bull case is real: if EU competition authorities find what Sabre is describing — airlines constrained, not choosing — forced interoperability at the PSS layer would open the O&O technology market in a way commercial competition hasn't achieved. Airlines with no practical path to Sabre's OSD or Accelya's FLX would get a genuine option. The bear case is the evidence problem: every public Nevio confirmation since Sabre filed — Air France-KLM, British Airways, Saudia — is an airline appearing to choose voluntarily, which is the opposite of the coercion regulators need to see. But the deeper point is structural. The PSS has become the modern distribution chokepoint the way the GDS was for thirty years. The difference: the GDS held data. The PSS runs the airline. You cannot build a parallel pipe around the system that processes your check-in, your inventory, and your departure control. The chokepoint moved. The incumbency advantage moved with it.
Model Context Protocol — the AI-to-data standard the tech industry has converged on — landed in airline distribution this week. TPConnects added MCP layers to its Iris and Astra platforms, exposing more than 60 airlines through a single machine-readable interface and normalizing NDC schema fragmentation for AI agents. The bull case: MCP solves the problem NDC created. Airlines built NDC to own their offer but implemented it differently at every carrier. An MCP normalization layer gives AI agents one clean interface. If agents become the dominant shopping front end — IDC projects 30% of bookings through agents by 2030 — airlines reach travelers without paying a new intermediary at the search layer. The bear case: a normalization layer over 60 airlines is still an aggregator. Whoever controls that layer controls what agents see, what they prioritize, and how airline content ranks. Airlines that spent a decade pulling offer creation in-house could find themselves handing the demand interface to whoever owns the best agent connection. The critical take: agents do not care about brand, bundles, or merchandising strategy. They return the math. Everything airlines built modern retailing to enable — rich content, dynamic offers, experiential ancillaries — gets compressed by an agent into a row in a comparison table. The MCP layer is not just plumbing. It is the moment airlines discover whether their offer survives being read by a machine.
The IATA AGM in Rio delivered the most substantive Offer and Order progress report the program has ever produced — and a timeline that demands honesty. Seven full-scope airline O&O contracts now in place. Forty-eight pilots or proofs of concept. IT-provider tenders doubling year on year. Riyadh Air live as the first full-service greenfield carrier on native O&O. TMCs reporting NDC bookings up more than 150% year over year. Qantas and the ATMC held a transparency forum on NDC adoption in Sydney on June 2. The bull case: these numbers are not slideware. The shift from experimentation to material volume is visible, and airlines that move now retire legacy cost and own their offer while competitors are still scoping. The bear case: seven full-scope contracts across an industry of 300 to 400 airlines is a thin base, most IT providers don't finish current builds until 2027 or 2028, and the IATA roadmap quietly acknowledges the full transition runs past 2030 for most carriers. The critical take: the winners in this transition will not be whoever reaches 100% Offer and Order first. They will be whoever runs the messy middle most cheaply — managing NDC plus EDIFACT plus O&O simultaneously, for the rest of this decade, without the cost and complexity eating the retailing upside. Coexistence is not a transition problem. It is the product.
Stories Referenced in This Episode
Narrative 1 — Sabre vs. Amadeus: PSS as the New Chokepoint
- Sabre Claims Amadeus Blocks Competition in Airline Technology — Skift, May 7, 2026
Narrative 2 — The MCP Layer Arrives
- MCP vs. NDC and other challenges facing airline distribution — PhocusWire, 2026
- Iris MCP Layer Enables AI Airline Distribution — TPConnects, 2026
- MCP Explained: The AI Standard Reshaping Travel Tech — Skift, December 2025
Narrative 3 — IATA AGM Rio: Real Momentum, Honest Timeline
- Progress on the journey to 100% Offers and Orders — IATA, December 18, 2025
- Modern Airline Retailing Industry Transition Roadmap — IATA, 2026
- Qantas and ATMC Forge New Path on NDC — Travel and Tour World, June 2, 2026
The Bottom Line
Airlines won the right to own their offer. They immediately found new gatekeepers forming above and below them. The PSS incumbency layer — which Sabre named publicly but hasn't yet been able to dislodge commercially or legally — is the chokepoint that determines whether O&O transformation is genuinely competitive or effectively pre-decided. The MCP normalization layer is building a new demand interface for the agentic era, and it's forming outside airline standards bodies without airline input. And the IATA program in Rio confirmed what the roadmap has said quietly for two years: the industry lives in hybrid coexistence until the early 2030s. The three floors of the same building. The airline that runs the messy middle cheaply — managing three distribution stacks, staying visible to agents, and actually deciding what to sell through the system it's building — wins the decade. Not the one that crosses the finish line first.
Intro music: The perfect corporate podcast intro by Lundstroem. Licensed under a Attribution 4.0 International License.
The V1 Airline Retailing Report publishes every Monday. Subscribe on Apple Podcasts, Spotify, or wherever you listen.
V1 Advisory LLC | v1advisory.co
Related: Tasks | V1-Advisory-Status | V1-Airline-Retailing-Report-Podcast-Guide
Chapter 1
Imported Transcript
Eric Marketts
Last week we closed with two specific instructions. Watch Rio — the IATA AGM, running June 6 to 8 — for any signal the industry is addressing the agent question directly. And watch whether Sabre's monopoly claim gets a regulatory hearing or quietly becomes a negotiating posture. We said next week we'd know which was signal and which was noise .
Steph Nell
This week the answer is: both are signal. Just not in the way Sabre intended.
Eric Marketts
Welcome back to The V1 Airline Retailing Report and today is Monday, June 8th. I'm Eric Marketts.
Steph Nell
I'm Steph Nell. Three narratives today and they share a floor plan. The PSS-and-offer-engine layer. The MCP agent interface forming above it. And the IATA program tying it all together in Rio. Three floors of the same building. And the question running through all of them is the same one: airlines spent a decade fighting to own their offer. Who's standing between them and the buyer right now?
Eric Marketts
Let's build it from the ground up .
Eric Marketts
We've covered the Sabre-Amadeus dispute twice — first in Episode 4, when Sabre CEO Kurt Ekert named Amadeus's Altéa lock-in on an earnings call, and again in Episode 6, when we showed how every new Nevio confirmation made the legal argument harder to sustain. This week the question isn't whether Sabre is right about the dynamic. It's what naming it actually reveals about where the real chokepoint is .
Steph Nell
Because Sabre is right about the dynamic. Walk through what that means precisely.
Eric Marketts
For twenty years the GDS was the gatekeeper between the airline and the agency. Not by natural market choice — because the GDS held the pipe. NDC was built to challenge that directly. Airlines would publish offers themselves, own the content, own the pricing logic. They won that argument. The standard exists. The pipes are being built.
Steph Nell
And in the Offer and Order world the offer engine bolted onto the PSS becomes the new gatekeeper. Not the distribution pipe — the operational system the airline runs on. Whoever built that system controls what transitions the airline can afford and what it can't.
Eric Marketts
The bull case for Sabre's legal move is real. If a competition authority — the EU is the credible venue — forces API interoperability at the PSS layer, it opens the O&O technology market in a way the market hasn't achieved commercially. Airlines that today have no practical path to Sabre's OSD or Accelya's FLX get a genuine option. Market competition drives better products and lower costs. The 73% of airlines that haven't started O&O transformation get a less obstructed path .
Steph Nell
The bear case is sequencing and evidence. Amadeus has processed 40 million Nevio bookings. Air France-KLM, British Airways, and Saudia have confirmed Nevio publicly. Regulators build antitrust cases around coercion and foreclosure. Voluntary confirmations are the opposite of that evidence base. EU investigations run two to four years, during which Amadeus keeps converting Altéa carriers and deepening switching costs every quarter .
Eric Marketts
Here's the critical take — and it's not about either company specifically. The PSS has become the modern distribution chokepoint the way the GDS was for thirty years. The difference is that this chokepoint is operational, not commercial. The GDS held data. The PSS runs the airline. You can build a parallel pipe to bypass a data intermediary. You cannot build a parallel pipe to bypass the system that runs your check-in, your inventory, and your departure control. Every NDC and O&O discussion that treats distribution as the variable and the PSS as the constant is looking at the wrong layer. The PSS is the constant. It always has been .
Steph Nell
Second narrative. And this one connects to something Sabre actually built — not the lawsuit. The Sabre-Mindtrip-PayPal agentic booking pipeline we covered in Episode 5 wasn't just a proof of concept. It was the first implementation of a pattern the industry is now building toward systematically. This week that pattern has a name and a standard: Model Context Protocol .
Eric Marketts
MCP. The AI-to-data interface that the tech industry has largely converged on as the way AI agents connect to external systems. It landed in airline distribution in two meaningful ways this week. TPConnects added MCP layers to its Iris and Astra platforms — exposing more than 60 airlines through a single machine-readable interface and normalizing NDC schema fragmentation for AI agents. Sabre, PayPal, and MindTrip separately announced what they're calling the industry's first end-to-end agentic booking pipeline targeting Q2 2026. The shared pitch: if your content is not machine-readable to an agent, you are invisible in agent-led shopping .
Steph Nell
The bull case is that MCP solves the problem NDC created. Airlines built NDC to own the offer — but every carrier implemented the standard differently. Different schema structures. Different ancillary representations. An AI agent querying 60 airlines through 60 NDC implementations is not practical at scale. A normalization layer hands agents one clean interface. If agents become the dominant shopping front end — and IDC's 30% by 2030 projection says they're heading there — airlines reach travelers without paying for a new intermediary at the search layer .
Eric Marketts
And the bear case is that an MCP normalization layer over 60 airlines is still an aggregator. TPConnects sits between the airline and the agent. Whoever controls that normalization layer controls what the agent sees, what it prioritizes, and what schema elements it surfaces. Airlines that spent a decade pulling offer creation in-house could hand the demand interface to whoever owns the best agent connection. New intermediary. Different name. Same structural problem .
Steph Nell
What's the critical take?
Eric Marketts
Agents do not care about your brand. They do not care about your bundle design, your loyalty pitch, or your merchandising strategy. They care about a clean, fast, machine-readable offer and a price. Everything airlines built modern retailing to enable — rich content, dynamic bundles, experiential ancillaries, brand differentiation — gets compressed by an agent into a row in a comparison. The MCP layer is not just distribution plumbing. It is the moment the airline discovers whether its offer can survive being read by a machine that strips out the marketing and returns the math .
Steph Nell
Third narrative. The 82nd IATA AGM and World Air Transport Summit, June 6 to 8 in Rio, hosted by LATAM. Modern Airline Retailing on the agenda. The momentum numbers coming out of Rio are the best the program has ever produced. They're worth reading carefully .
Eric Marketts
Seven full-scope airline Offer and Order contracts in place industry-wide. Forty-eight pilots or proofs of concept. IT-provider tenders doubling year on year. Riyadh Air live on native O&O as the first full-service greenfield carrier. TMCs reporting NDC bookings up more than 150% year over year. And Qantas and the Australian Travel Management Company held a transparency forum on NDC adoption in Sydney on June 2 — notable because Qantas hasn't historically been aggressive on NDC transparency .
Steph Nell
Bull case. These numbers are not slideware. Seven full-scope contracts. Doubling tenders. A greenfield carrier on native O&O in production. Triple-digit NDC booking growth. The shift from experimentation to material volume is in the data, not just in conference announcements. Airlines that move now retire legacy distribution cost, own their offer, and merchandise dynamically — while competitors are still scoping programs.
Eric Marketts
Bear case: read the roadmap, not just the momentum. Seven full-scope contracts across an industry of 300 to 400 airlines is a thin base. Most IT providers completing current builds don't finish until 2027 or 2028 — and PSS decommissioning, the hard part, begins only after that. Triple-digit NDC growth off a small base is mathematically easy. The IATA roadmap itself quietly acknowledges the full transition runs past 2030 for most carriers. The visible momentum is tenders and pilots. Production at scale hasn't started .
Steph Nell
And the critical take ties all three narratives together. The transition runs past 2030. Which means the industry lives in hybrid — NDC plus EDIFACT plus O&O — for the rest of this decade. The PSS chokepoint from narrative one and the MCP agent layer from narrative two are both forming during that hybrid period. The winners in this transition will not be whoever reaches 100% Offer and Order first. They'll be whoever runs the messy middle most cheaply — managing three distribution stacks at once without the cost and complexity eating the retailing upside. Coexistence is not a transition problem. It's the product for the rest of the decade .
Eric Marketts
Three narratives. One question. Let me close it.
Steph Nell
The MCP layer is forming outside airline standards bodies, outside IATA's process, and outside any certification framework. It's being built by the tech industry to solve a tech-industry problem — how AI agents access structured data at scale. Airlines are not at the table where MCP architecture is being decided. That is not a hypothetical gap. It is a gap that is closing, one implementation at a time, without airline input .
Eric Marketts
And Rio gave us the honest version of the transition timeline. Real momentum. Real contracts. Real volume growth. And a roadmap that acknowledges most of the industry won't finish until the early 2030s. The urgency isn't the destination. It's the decade of coexistence ahead — and whether you're managing that cost or being managed by it.
Steph Nell
The PSS layer is the incumbent gatekeeper. The MCP layer is the emerging one. And the industry program runs through both of them at whatever speed the slowest PSS migration allows. Same building. Three floors. Built in the wrong order. If this episode landed — share it with someone in your organization who's making distribution or retailing decisions in the next 12 months. These are not 2028 questions. They're 2026 questions with 2030 answers.
Eric Marketts
We'll be back next Monday. I'm Eric Marketts. Also, if you want to dig deeper into these topics please see the episode details for specific links and news articles.
Steph Nell
I'm Steph Nell. Thanks for listening.
Eric Marketts
Stay sharp out there .
