Every few weeks another airline announces that its NDC content is now “live”. The press release follows a familiar script: a new distribution partnership, a market count, a quote about richer content and personalised offers, and the implicit suggestion that another airline has crossed the finish line.
It is not. For airlines, millions of dollars in technology investment depend on that distinction. For agencies, it determines whether a ticket can actually be serviced after it is sold. In a recent industry survey, 81% of airline respondents said they have live NDC channels, with some carriers exceeding 50% NDC penetration of indirect bookings. That number, on its own, tells you almost nothing about whether NDC is actually working for those airlines, their agents, or their travellers. It tells you that a channel exists. Whether it carries meaningful volume, whether agents can service what they sell through it, and whether the airline’s own systems can absorb the cost of running it, are three separate questions that “live” does not answer.
What “live” is actually certifying
IATA built a certification structure precisely because “live” had become too loose a word to mean anything on its own. NDC Level 4 covers implementation with full offer and order management, including shopping, booking and a limited range of post-booking processes such as changes and cancellations, while NDC@Scale evaluates a carrier or seller across technical setup, organisational setup, use cases and capabilities. These are meaningful thresholds. But they measure capability, not volume, and certainly not commercial outcome. An airline can be fully certified and still be pushing a low single-digit share of its indirect bookings through the channel.
United Airlines’ own numbers make the point better than any outside critique could. Speaking at Airline Distribution 2026, the carrier’s managing director of digital sales described a channel mix that is roughly 70% direct, 10% NDC, and 20% EDIFACT. That is one of the more digitally mature carriers in the world, years into its NDC programme, still carrying a fifth of its business on legacy messaging. “Live” happened for United a long time ago. “Working,” in the sense of NDC displacing legacy distribution at scale, is still an ongoing and incomplete process even for a carrier of that size and technical sophistication.
The bill that arrives after the switch flips
Part of why so many NDC programmes stall between announcement and production maturity is that going live changes the cost structure before it changes the revenue. Under the old GDS model, the distribution intermediary absorbed the computing cost of search traffic. Under NDC, the airline does. As TPConnects has argued, implementing NDC requires airlines to re-architect core distribution systems while continuing to operate their legacy infrastructure in production. Airlines are not building this in a lab. They are building it while continuing to sell tickets on the systems the new one is meant to replace.
The look-to-book ratio is where this shows up most starkly. For large OTAs pushing differentiated content, that ratio can reach 20,000 to one, meaning resellers make twenty thousand API calls to flight inventory providers for every ticket sold. Under the GDS model this was somebody else’s server bill. Under NDC, it is the airline’s. Going live is therefore the moment an airline begins paying the infrastructure costs of a channel whose commercial return has yet to be proven.
Even when NDC reaches meaningful booking volumes, airlines still have to demonstrate that those bookings improve economics. Lower distribution costs can be offset by higher infrastructure spending, increased API traffic, greater servicing complexity, and ongoing investment in modern retailing capabilities. Ultimately, the success of an NDC programme is measured not only by adoption, but by whether it produces a better financial outcome than the model it replaces.
The agent is where “working” gets decided
None of this reaches the traveller directly. It reaches the travel agent first, and the agent is where the gap between certification and function is most visible. A panel at WiT’s Africa pop-up series concluded that the industry’s original ambition of a single plug-and-play NDC integration has not materialised. Instead, each airline has developed its own customised implementation, undoing much of the simplification NDC was meant to deliver, forcing agents to integrate each carrier separately. The same panel identified post-sale servicing, particularly changes, refunds and special requests, as the single biggest pain point, with no standard servicing logic across carriers.
This is the detail that “NDC is live” announcements almost never mention, because it is not the airline’s problem to announce. It is the agent’s problem to absorb. A channel can be technically live and commercially unusable at the same time, if the agent booking through it cannot confidently handle what happens after the ticket is issued.
Where this bites hardest
TDN has covered this gap directly on the African continent, where the gap between technical readiness and commercial reality is often widest. AFRAA Secretary General Abderahmane Berthé has been direct about it: the NDC standard, in his assessment, does not yet adequately serve the majority of African carriers, and the survey data he cites shows only eighteen percent of carriers plan to complete their NDC transition within a year, with a further twenty-eight percent expecting to do so within two to three years. Berthé’s diagnosis is not that African carriers are behind on ambition. It is that the commercial and technical models built around NDC assume a scale and capital base that most of the continent’s carriers do not have, and that the travel agent community remains far more embedded in GDS workflows than technology vendors arriving on the continent tend to assume.
TDN’s own reporting on African travel agencies found the same pattern from the seller’s side of the desk. Each carrier’s differing NDC maturity, aggregator partnerships, surcharge structures and servicing standards leaves agents managing what is effectively a separate relationship per airline, with none of the promised simplification. The surcharges tied to legacy GDS bookings are already live. The content divergence between GDS and NDC channels is already happening. For an under-resourced agency, the industry’s rollout of “live” NDC channels can function less like modernisation and more like a slow-motion reduction of the content they are able to sell competitively.
From proving the concept to making it normal
There is a more optimistic reading available, and it is worth stating plainly rather than dismissing. Industry observers going into 2026 describe the shift as one from proving the concept to making it normal, with pilots turning into production-grade rollouts and API performance finally becoming boring rather than newsworthy. That is a real and welcome maturation. Singapore Airlines is regularly cited as the reference case for how to do this properly. Rather than treating NDC as a single “we are live” moment followed by silence, the carrier treated it as an organisational transformation rather than a technology deployment, backing that up with parallel systems during the transition and continuous investment in trade education.
The distinction TDN wants to draw is not that NDC is failing. It is that “live,” “working,” and “profitable” are three different claims, and the industry’s press release culture has spent a decade collapsing them into one.
A certification is not a volume.
A volume is not a servicing capability.
A servicing capability in one market does not guarantee the same outcome in another.
And none of the three guarantees the fourth thing that ultimately matters: a better financial outcome than the model it replaced. What holds at a flagship carrier’s home market does not hold at a smaller carrier with a fraction of the technical budget, or at an agency where the GDS relationship still pays the bills. Every one of those gaps is where an NDC programme can be genuinely live and genuinely not working, at the same time, for years.
The carriers and technology partners who will look credible in five years are unlikely to be the ones with the loudest “NDC is live” announcements. They will be the ones who can say, with evidence, what share of bookings actually moved, what happened when a traveller needed a refund six weeks later, and whether the agent who sold the ticket could handle that without calling a help desk that did not know the answer either.
TDN will continue tracking NDC production milestones across its core African and Middle Eastern markets, distinguishing certification announcements from verified adoption data where the two diverge.
