Praveen Kumar, Co-Founder of TPConnects: “While We’re Speaking, the Industry is Moving Towards ONE Order”

Q: You’ve been involved in NDC since its early days. Why do many pioneers like British Airways or American Airlines still operate on version 17.2? 

Praveen: The main reason is simple — 17.2 was the first truly stable baseline for the industry. Earlier versions, such as 15.2 and 16.2, were not centered on the Offer and Order model. Version 17.2 changed that by establishing a clearer structure, along with order management and servicing workflows.  Airlines built a significant amount of commercial logic and integrations around 17.2. Over time, moving away from it became increasingly complex because it requires transformation across the broader ecosystem, which is why many carriers still remain on this version today.

Q: IATA supports 13 NDC versions simultaneously. Should resellers care about these differences?

Praveen: Airlines not only use different NDC versions — they also implement them in their own way, layering custom workflows on top. As a result, resellers are forced to navigate inconsistent shopping processes, varying seat map structures, different approaches to ancillaries — such as weight-based versus piece-based baggage — and fragmented servicing logic. That’s why aggregators play such a critical role — they act as a normalization layer, translating highly fragmented airline content into a consistent and usable experience.

Q: Looking back, which NDC releases were real breakthroughs?

Praveen: If I had to highlight the most important ones, 17.2 stands out as the true starting point because it established the Offer and Order foundation and formalized servicing workflows. Version 20.2 took this further by extending the scope beyond retailing into accounting and delivery. Then 21.3 emerged as a stable, widely adopted baseline that maintains backward compatibility. More recently, the 24.1 version brought together offer, order, servicing, settlement, and disruption handling into a more unified model. So, it’s going to be another next-gen NDC breakthrough.

Q: What are the other key priorities of version 24.1 and beyond?

Praveen: One major area is digital identity, which allows all parties in the distribution chain — suppliers, retailers, aggregators, and agents — to be clearly identified and authenticated. There are also improvements in payment handling, settlement logic, and servicing capabilities, along with better support for disruptions and order lifecycle management. Overall, the direction is toward a more connected and transparent ecosystem.

But it’s important to understand that the most recent versions are built on top of 21.3. So even when an airline adopts 24.1 or newer versions, they are effectively working from that baseline. The newer releases extend and refine the capabilities rather than replacing them entirely.

Q: What actually pushes airlines to upgrade their NDC versions?

Praveen: It usually comes down to a combination of factors. Airlines need better servicing performance, especially for real-time changes and repricing. They also want to unlock more commercial opportunities through ancillaries and personalization. At the same time, they have to align with their technology roadmaps and move toward Offer and Order models. And importantly, there’s pressure from partners — aggregators, TMCs, and corporate clients — who expect more advanced capabilities.

Q: Many airlines don’t fully migrate; instead, they mix versions. Is that a problem?

Praveen: It’s not ideal, but it’s a practical approach during transition. Often, airlines use newer versions for booking while keeping servicing on older versions like 17.2. This creates complexity for resellers, who have to deal with multiple APIs and workflows, but it reflects the reality of how difficult full migration is. Over time, these hybrid setups will likely converge into more unified implementations.

Q: What NDC capabilities are still underused today?

Praveen: Ancillaries are a big one. There’s a lot of potential for airlines to expand beyond their own products and dynamically integrate third-party services, but that’s still underutilized. Another area is servicing automation — things like refunds, waivers, and notifications are supported in the standards, but not consistently implemented across the industry.

Q: NDC still struggles with interlining. Why is that?

Praveen: So today, NDC mostly lives in the retail space. Interlining is a bit different — it’s really about one airline acting as a retailer for another. And that comes with a lot of layers: contracts, rate parity, settlement… all of that.

Until around 20.2, there wasn’t really a consistent framework in NDC to handle this. But now, especially with 21.3, we’re seeing a shift from what used to be a loose meta-agreement into a more structured seller–retailer model.

In that setup, airlines interact directly via APIs instead of relying on EDIFACT. So one airline can call another in real time and ask: is this available, what’s the price, can I sell it? And based on that, the retailing airline builds the full offer on behalf of all suppliers.

What’s important here is that you don’t have to wait for all the traditional post-processing — things like proration through clearing systems. In a one-to-one API connection, the revenue split is already defined at the moment of sale.

Of course, reworking GDS-driven processes that have been around for decades takes time. But there is a clear direction from IATA, and from what I’ve seen in discussions and workshops, the industry is definitely moving that way.

Q: Where is the NDC schema today in terms of ONE Order?

Praveen: So 18.2 laid the foundation for ONE Order, 20.2 extended it into accounting and delivery, and 24.4 brings a much higher level of maturity. We’re now seeing a new set of messages around order-based servicing and status updates, moving away from traditional PNLs (Passenger Name Lists) and separate ancillary or check-in communications toward everything being handled within the Order and its items.

As this evolves, all references shift to the order item level. Each item can include multiple service products, and each service has its own delivery tracking — so there’s no longer a need for tickets or EMDs. This also opens the door for better ancillary sales, including in interline and codeshare scenarios.

In the end, ONE Order brings everything into a single flow — from offer and order creation through check-in and baggage handling — all within one data model. And with 24.4, a lot of this is already defined, so we can expect wider adoption in the coming years.

Q: If the standards are already in place, what are the main challenges preventing airlines from moving to ONE Order in practice?

Praveen: Here we’re not talking about one or two platforms, but the entire ecosystem. This is a full industry transformation, and it’s not something that can happen within a year or two.

If you look at TPConnects, we’ve worked closely with providers such as Maureva, Lufthansa Systems, and other delivery platforms to implement ONE Order capabilities. At the same time, there are still many constraints in downstream systems that need to adapt to this new model. But that is gradually changing — and while we’re speaking, the industry is already moving toward ONE Order.

Q: So when can we expect ONE Order to reach widespread adoption?

Praveen: If we look at history, it took about a decade to fully adopt e-tickets. Given how technology is evolving today, I would expect this timeline to be shorter, perhaps bringing widespread ONE Order adoption into the 2028 to 2030 timeframe. The most critical piece will be interlining, which will likely be resolved toward the later stages of that period. Overall, the next five years are going to be very interesting for the industry.

Q: How do you see airline retailing evolving?

Praveen: The focus will shift to how personalized the offer is — how it adapts to each seller, each frequent flyer, and how much value it brings through ancillaries using dynamic bundling and dynamic price optimization.

This is where the industry is heading — toward dynamically generated offers and orders that are consistently communicated across the infrastructure, including departure control systems and accounting platforms. At the same time, technologies like digital identity and biometrics will become part of the broader travel experience.

Ultimately, the future is not about which NDC version you use. It’s about how effectively you can generate offers, manage orders, and deliver services across the entire distribution ecosystem in real time.