TL;DR: NDC isn't replacing EDIFACT on any predictable near-term schedule — even the most advanced airlines are still in an early setup phase for full Offer & Order (O&O) capability, and industry analysis puts mainstream O&O adoption no earlier than 2030. For travel platforms, that means building for a genuinely hybrid distribution environment for years, not planning a clean cutover. This guide covers where NDC and EDIFACT actually differ, why airlines implement NDC inconsistently despite a shared standard, what integration actually costs today, and how to architect booking software that handles both without forcing a rebuild when the balance shifts.

Why this transition is taking so much longer than announced

New Distribution Capability (NDC) was introduced by IATA over a decade ago as a modern, XML/API-based replacement for EDIFACT, the decades-old text-based messaging standard that still underpins most airline distribution through GDS platforms. The promise was richer content (ancillaries, personalized offers, dynamic pricing) and a move from rigid, message-based distribution to something closer to modern e-commerce. As of 2026, more than 70 airlines are NDC-certified according to IATA, with major carriers including American, United, Lufthansa Group, and British Airways now routing premium content exclusively through NDC channels.

But certification and full replacement are different things. Airlines remain deeply invested in EDIFACT infrastructure, and migrating away from it involves real costs in training, system replacement, and operational risk that most carriers are managing cautiously rather than rushing. Even the leading NDC airlines are still in an early setup phase for full Offer & Order capability — the deeper order-management functionality that lets NDC handle the complete booking lifecycle, not just search and initial booking — while many other airlines won't begin that transition until 2028-2029. Current industry analysis suggests mainstream Offer & Order adoption is unlikely before 2030, which is also roughly when legacy passenger service systems and EDIFACT messaging are expected to start being meaningfully phased out, not eliminated overnight.

For any travel platform — an OTA, a corporate travel management company, or a custom booking engine — the practical implication is unambiguous: you're building for a multi-year period where both standards coexist, not planning around a single migration deadline.

NDC's technical promise creates its own operational problems

NDC's shift toward richer, more dynamic content has introduced a genuinely new operational challenge: a dramatic increase in the look-to-book ratio, the number of search and shopping requests airlines process relative to actual bookings. NDC has pushed this ratio from a few hundred to thousands, and AI-driven search and comparison tools are amplifying that pressure further through higher automated query volume. This isn't just an airline infrastructure problem — it raises a real business model question about who bears the cost of that query volume, and it's part of why some airlines are more cautious about broad NDC rollout than the technology's maturity alone would suggest.

Standardization is the other recurring friction point. Even though NDC is a shared IATA standard, different airlines implement it slightly differently — a familiar pattern from other API standardization efforts, where a common specification doesn't guarantee common implementation. A travel platform integrating multiple NDC-enabled airlines is still integrating multiple, meaningfully different data models and booking flows, not one uniform API.

What NDC actually adds over EDIFACT — and where EDIFACT still wins

NDC's core advantages are real: richer product content including branded fares and ancillaries, more dynamic and personalized pricing, and an architecture built for modern API-based integration rather than fixed-format text messages. For airlines, this supports genuine retailing — presenting offers the way an e-commerce site would, rather than a fixed inventory of fare classes.

EDIFACT's advantage, at this point, is simply maturity and ubiquity. It's a known quantity, deeply embedded in GDS infrastructure (Amadeus, Sabre, Travelport), well understood by travel agency booking tools, and functional for the vast majority of standard itinerary bookings that don't require NDC's richer content model. For a platform that doesn't need branded fares or dynamic ancillary bundling as a core feature, EDIFACT remains a perfectly workable distribution channel — and will continue to be for years, given the adoption timeline above.

How most platforms actually access NDC content today

Very few travel agencies or booking platforms connect directly to each airline's individual NDC API — the integration overhead of maintaining separate connections to dozens of airlines, each with its own implementation quirks, doesn't scale. Instead, most access NDC content through aggregators: GDS platforms like Amadeus, Sabre, and Travelport now offer NDC content alongside their traditional EDIFACT feeds, or specialist NDC aggregators like Travelfusion normalize NDC content from multiple airlines into a more consistent interface.

This aggregator-first pattern is becoming the de facto standard specifically because of the normalization problem: modern distribution platforms increasingly carry EDIFACT and NDC content in parallel, presenting both within a single shopping and booking experience for the end user, while the aggregator absorbs much of the burden of reconciling different airlines' inconsistent NDC implementations. In practice, this means most NDC work sits on top of existing GDS integration rather than replacing it — the GDS connection stays in place for EDIFACT content while adding an NDC content stream alongside it.

What NDC integration actually costs in 2026

For a platform integrating a single airline's NDC API directly, cost typically runs $8,000 to $50,000+, depending on the specific airline's API maturity, the booking flow being built, platform type, and how much automation and certification work is required. Multi-airline platforms face meaningfully higher costs — typically $50,000 to $100,000+ — because every additional airline can follow a different NDC version, offer structure, pricing flow, and order-servicing model, and normalizing across that inconsistency is real backend engineering work, not configuration. That normalization layer alone, for a multi-airline platform, typically runs an additional $15,000-$40,000+.

Beyond the initial booking integration, a functional admin panel for managing NDC bookings, customers, agents, payments, cancellations, refunds, and reporting adds further cost, and that cost rises further with advanced analytics, role-based or branch-wise access, and supplier-specific reporting. Mobile app integration adds cost on top of that, since the NDC booking flow needs to work reliably across platforms while staying connected to a secure backend handling real payment and order data.

Architecture principles for a multi-year hybrid period

Given that both standards will coexist for years, the platforms that handle this well share a few architectural characteristics:

A unified internal offer and order model, not two parallel code paths. Rather than building separate booking flows for NDC and EDIFACT content, the more resilient approach normalizes both into one internal representation of an offer and an order as early as possible in the pipeline — every downstream feature (search, pricing display, booking, servicing) works against that unified model, regardless of which distribution channel the content originated from.

Per-airline configuration, not a single NDC integration. Because different airlines implement NDC differently, the integration layer needs per-airline adapters or configuration, similar in principle to how a resilient open banking or channel manager integration handles per-institution quirks rather than assuming one implementation generalizes to all.

Aggregator-first, with direct integration reserved for genuine need. For most platforms, integrating through a GDS's NDC offering or a specialist aggregator is the more sustainable starting point — direct airline-by-airline NDC integration is worth the added cost only when a specific airline's content, pricing, or servicing capability genuinely isn't available (or isn't reliable) through existing aggregator relationships.

Order management built for the full lifecycle, not just booking. The strongest growth in verified NDC capabilities industry-wide is specifically in payments and order servicing — the parts of the booking lifecycle beyond initial search and booking that make NDC genuinely usable end-to-end. A platform that treats NDC integration as "booking only" and bolts on servicing, cancellations, and refunds later is setting up for a second integration project rather than one done right the first time.

Monitoring for look-to-book ratio and query cost. Given NDC's dramatically higher look-to-book ratios, platforms doing significant NDC search volume need to monitor and manage query patterns deliberately — both to control cost and because some airlines apply their own rate limits or commercial terms tied to query volume.

Building for the transition, not a deadline

The travel platforms handling this well aren't the ones betting on a specific NDC cutover date — they're the ones that architected for a multi-year hybrid distribution environment from the start: a unified internal data model, per-airline configuration where needed, aggregator-first integration, and order management built for the full booking lifecycle rather than just the initial sale.

Syslabs works with travel platforms on exactly this kind of custom booking software and NDC integration work — from aggregator-based implementations to direct airline connections and the order management layer underneath both. If you're scoping an NDC integration or rearchitecting a booking platform to handle both standards cleanly, a short technical review is usually enough to identify where the real complexity sits.

Sources

  • AltexSoft, OAG, PhocusWire, Tragento, OneClickITSolution, SlideServe, AirlineDistribution.net — 2026 NDC vs. EDIFACT transition and airline distribution analysis
  • AltexSoft, Track360, Navan, OAG, The Travel Team, Travel Code, Forecepts, NDC Deals — NDC adoption timelines and airline certification data
  • GuruTechnolabs, Booking.com for Business, United.business, Silvi Global Technology, Trawex, FoundersWorkshop, FlightsLogic — NDC integration cost and travel booking software development benchmarks