google flights iata ndc

Might Google benefit the most from IATA’s NDC?

NB: This is a viewpoint from Paul Richer, owner of travel technology consultancy Genesys.

Last month I chaired the Travel Technology Initiative’s Summer Forum – Take Flight!, which covered developments in flight distribution.

Speakers discussed Google Flights, among other topics. There was also a presentation about next-generation fare content distribution by Jerry Foran, head of product delivery, revenue management, British Airways.

Foran’s latter drew a lot of attention. IATA’s New Distribution Capability (NDC) is a brave initiative by the airlines to bring flight distribution into the 21st century.

air canada ndc

As Foran pointed out, right now airlines file fares and schedules with third parties, GDSs package and push offers based on third-party databases (price/frequencies), agents submit travellers’ requests using GDS and the airlines are the last to know who has purchased airline tickets.

This way of doing business was no doubt considered to be perfectly satisfactory in the 1970s and 1980s. However, airlines have reached a stage of sales sophistication with their own websites that outstrips the sales capability of many of the third party channels that distribute their products.

As Foran mentioned, airlines can deliver rich content on their websites, bundle offers clearly stating what is included in each and provide consumers with the ability to create their own offers–choosing meals, selecting and paying for preferred seating, paying for additional baggage, and so on.

google flights iata ndc
The Google question

A key point, raised afterward by Milind Puri of Kuoni, is who really is going to invest in point-to-point connections with the various NDC-compatible airlines.

Google is one company that came to my mind.

deep link google flight search

That thought brings me to one of the other presenters at the forum: Massimo (Max) Morin, head of technical sales EMEA & APAC for Google. Max was technical consultant lead at ITA Software up until it was acquired by Google.

Just to remind you of the significance of this acquisition in Google’s own words:

“Google is buying ITA Software to create a new, easier way for users to find better flight information online. By combining ITA Software’s expertise with Google’s technology, we will be able to build new flight search tools for users that will make it easier for them to search for flights, compare flight options and prices, and get them quickly to sites where they can buy their tickets.”

So Google Flights was born as a way of helping travellers with simple, easy access to flight options for their journeys. It is easy and intuitive to use, blazingly fast but lacks the all important ingredient for travellers, consistently competitive fares.

I tried a couple of flight searches on Google Flights and compared the fares displayed to those available from several online travel agents. Prices displayed were undercut on every occasion.

But imagine a future a few years from now where Google Flights could be able to use NDC to connect to many airlines and receive up-to-the-minute flight availability and pricing.

Think of a scenario where Google, with its massive user base, offers airlines the opportunity to participate in Google Flights at a cost lower than airlines’ GDS/travel agent distribution costs.

If you were in charge of your airline’s distribution, would you not jump at the chance to be part of Google Flights? Not only would you be lowering your distribution costs but Google Flights would be passing travellers directly to your website where you could then regale them with enticing upsell offers.

Is this what the GDSs and travel agents fear?

NDC could be good for Google, technologically

NDC access would presumably be good for Google, as it would essentially be able to plug yield (availability) information right into its servers, rather than using the patchwork solution that it has today.

The NDC is much needed

These thoughts about Google beg the whole question about the NDC initiative in the first place. At its core, it’s about an update in messaging standards for fare content. But implicit in NDC is the belief by the airlines that the messaging medium that conveys their product to GDSs and agents is no longer fit for purpose. It is too basic.

So under the auspices of IATA they are developing a sophisticated XML message set that will address today’s deficiencies.

Of course, the middlemen – GDSs and travel agents – are not all supportive of this initiative. Some have stated they will embrace it but others are concerned that the distribution boat is being rocked and, I guess, they might be worried that they will fall overboard.

Jerry specifically addressed the five NDC myths that are being banded about:

  • NDC will result in less price transparency for consumers
. In fact, NDC will enable greater product, service and price transparency than the current filing system.
  • NDC will result in higher prices for consumers
. In reality increased competition usually results in lower prices.
  • NDC will result in less competition among airlines
  • There will be increased competition between airlines because they will be able to compete on service offerings as well as price.
  • NDC will prevent any comparison shopping
  • NDC will enable airlines to distribute all their products, not only the ones that fit into a GDS, and travel agent websites will be able to display and compare all those products.
  • NDC will violate passengers’ privacy rights in building personalised offers
o Passengers have the option to receive personalised offers. Airlines have to meet requirements of privacy regulations.

What is the ultimate worry of the naysayers about NDC?

One of the aspects of NDC is that it will provide a common messaging standard for flight resellers to use to connect to many airlines. This has to be good, doesn’t it?

Well, for agents and GDSs, perhaps not, because it will allow competitors to more easily enter the market.

Another issue is that while the GDSs’ technological infrastructure is geared-up for massive transaction volumes, airline systems connected to Google Flights and other intermediaries via NDC might simply not be able to handle the high search volumes.

Adjusting will require airlines to invest heavily in beefing up their technology and the cost of that might be prohibitive.

There are undoubtedly other concerns. What issues do you see as new, dynamic messaging standards are set to take off — with or without IATA’s help?

NB: This is a viewpoint from Paul Richer, owner of travel technology consultancy Genesys, where he blogs.

Image courtesy of Shutterstock; mashed up with Google logo. Screenshots from IATA presentation materials on the NDC and Google ITA Software presentation.

Share on FacebookTweet about this on TwitterShare on LinkedInEmail to someone

About the Writer :: Viewpoints

A founding principle of tnooz was a diversity of viewpoints from across the spectrum. Viewpoints are articles by guest contributors from around the travel and hospitality industries. The views expressed are those of the author. and do not necessarily reflect those of the author's employer, or tnooz and its partners.



Your email address will not be published. Required fields are marked *

  1. Glenn Wallace

    I was not trying to defend the status quo, I just wanted to point out that there has bee a lot of good collaboration put into it. I can think of many good examples, e-ticketing, private fares, optional services.

    I don’t understand your comment about wanting or needing NDC, I don’t know who you are so couldn’t possibly make that judgement.

    When I said “you’ll need it”, I meant the industry will need some good luck to sort through such a complex minefield.

  2. David Friderici

    I have to admit that I am really surprised that this topic (NDC) raises so many emotions and focusses so less on the objectives. But even more I am surprised to see obviously so many people/companies/agencies/etc. that feel threatened by NDC instead of starting to see real new opportunities and there are definitely many.
    Our planet is rotating and we should accept this as a given fact. NDC is introduces a new era. The entire industry, even our society and people’s behaviours has changed only in the last 5 years dramatically (Social media, mobile devices, etc.).
    Isn’t it logical that this also has consequences on industries that never encountered any significant structural changes for the last 30-40 years? Why do GDS fare displays and availability displays still look like 30 years ago? Why took it so long until an idea like ( has been realised? Because the core structures have never changed.
    All this now starts to happen and I am absolutely sure that nothing will stop this evolution as it is overdue.

    David Friderici
    VP & Head of Product Mgmt. and Strategy – Airline Travel Services at IBS

    • Glenn Wallace

      The best innovations come when the solutions address the needs of more members of the ecosystem. When I was working on PC based flight search at Expedia in 1999, the industry thought it was heretical to think it could be done on anything other than a mainframe. The guiding light for us was to show the consumer more selection, and better choice, and that would help consumers find what they wanted (at the time you could see maybe 3-5 options from the GDS) and we knew it would help suppliers by improving conversion and sales, and help us by growing our small (at the time) business. It also turned out to be very scalable.

      I have no vested interest in NDC succeeding or failing – I’m between jobs and doing some consulting – but I believe solutions should be evaluated on their merits. None of the airlines came to the industry and said “here are our goals, what can we do together to achieve them?”, they essentially took an existing solution and architecture from a vendor that had very limited market penetration and said “this is now our solution, like it or not”.

      They also didn’t come to the industry and say “what are your goals?” or “what do you think of our goals?”.

      What is tragic is that they there isn’t more creative and collaborative thinking going on.

      PLUS there is the tragic misdirection of “Why are you against modern XML standards!?” which of course no one is against. The disagreement is around the data flow, the semantic of the interface, and the overall architecture and structure.

      • David Friderici

        This is not what I observe. In none of the above posts anyone raised any doubts regarding the data, semantic or architecture. I even have the feeling that people complain without having any idea on how structures, interfaces and architectures look like. This would indeed be a much more exciting and valuable conversation. But there is not really a discussion about it.
        My observation are either really scared agencies because a lack of understanding and parties that simply still make much profit at lowest costs as long as everything remains as it was for the last decades. This is simply wrong.
        As said, a debate on the architecture would indeed be a valuable debate, but we do not have this debate here nor now anywhere else.

        • Glenn Wallace

          I’ve looked at all of the public NDC documents, I’ve seen the diagrams. I raised two critical points in the data flow, architecture and semantic of the messages:
          1) how scalable is a pull model vs a push model
          2) how and why can airlines do a better job selecting inventory to show consumers than 3rd party retailers and GDSs?

          No one has responded to either of those very fundamental architectural points, and continue to beat the drum of “why are you against new things” and “why are you scared” (ad hominem) without addressing any of the technical or consumer merits, or even talking about the actual goals.

          I haven’t seen any agencies or GDSs post here. I certainly have no emotion about it — my future may not even be in the travel industry. So I’m not sure who you are referring to.

          The agencies and GDSs are willing to help, but they want to be a part of the conversation. The GDSs and agencies have built an amazingly robust and flexible way of selling a great product, that the airlines have been refining and improving over the years. It has been a great collaboration. I think that collaboration is the path forward.

          I wish you all the best of luck with NDC. You’ll need it.

          • David Friderici

            When I was referring to agencies and GDSs I did this in general and was not referring to the discussion here only.
            When you refer to GDSs and agencies that have build an amazingly robust and flexible way of selling, I would not even disagree. The old Volkswagen Beetle is also an incredible robust reliable car.
            GDS fare and availability displays is probably not what you meant, when you used the terminology “flexible”. The low fare tools of the GDSs are also not what I would call state-of-the-art and I assume you would even agree. So what is it then that you think is worth to be defended?
            It is also hard for me to follow why you think that I or we need NDC. In case you referred to the fact that we are today one of the smaller PSS vendors give me the chance to add that in particular those PSS vendors that have new generation PSSs in place are certainly those ones who would have the fewest problems if this thing would not happen as all of them have their open interfaces already today in place … including available fare displays that do return more attributes than just an availability and a price sorted by price.

  3. Glenn Wallace

    > I must take issue with the assumption that a supplier controlled totally centralized mechanism is the only solution.

    That is what the NDC proposal is, from the public documents I have seen.

    > The current convoluted – fare component publish, multi-calculate, different answers – system fails to allow real time dynamic pricing in the neutral retailing channels.

    The real time dynamic pricing has been happening for years. Airlines choose which fare level they want to apply by opening and closing booking codes. If it doesn’t have enough granularity or personalization, that can be solved without throwing away all the things that work and scale. Because something is old and proven doesn’t mean it is bad.

    Also, there are dynamic pricing solutions for suppliers possible outside the “full content” world. And for a reasonable/smallish fee I can tell you how to do it. 🙂

    Perhaps an interesting question to ask is how a company like Google that is laser focused on instant response time, can take advantage of an architecture like NDC that requires off-host multiple queries to multiple suppliers, or implement a supplier personalized pricing/offer strategy without storing some data locally, be it pushed or pulled?

    > The rather bigger problem in my view is a commercial/political one.

    Actually I think it is a struggle over who selects which inventory to show the consumer. In the NDC world, the supplier does it. Can you imagine Samsung telling Amazon which Samsung products to show each and every customer?

    In the retailer/OTA world, the retailer comes up with what they think the customer will buy, and they continue to try to refine that right assortment.

    • John Pope

      The key highlight and takeaway of this recent exchange:

      “Can you imagine Samsung telling Amazon which Samsung products to show each and every customer?”

      All other, previously argued, details on this matter are really just secondary semantics.

      Game – Set – Match, Mr. Wallace.

      Well played, Glenn, well played.

      • Timothy O'Neil-Dunne

        Then clearly you playing Croquet and the rest of us are playing Rugby.

        Of course suppliers can determine their own product distribution. Samsung FIRST decides which channel it wants to distribute in and THEN decides if it will play by the rules of that channel. The argument goes both ways. Should Amazon be telling Samsung it can ONLY distribute a certain way and have pricing authority over Samsung? Of course not that is ridiculous.

        The point both of you seem to miss is that Amazon is (way) more sophisticated in its world than the current distribution channels are in Travel. It already offers personalization and Amazon determines who gets to see what in its own networks. It makes that available to direct distributors and other intermediaries alike along with a rule system to determine who gets what. I would also add that pricing a book or a physical product is more simplistic than the conditional requirements of a Travel product sale.

        Opening and closing fare buckets does not make for dynamic pricing. For example if a new fare of different value is filed and made available on a competitor’s website an airline has to wait for the fares to be filed and then distributed to the marketplace. Because the carrier signed a Full Content Agreement with the GDS technically if it publishes that fare first on the website before making it available via ATPCo powered GDS channels it would be in breach of that agreement.

        The consumer believes she/he is getting neutral choice and as Chris points out the decision about what to show a consumer is determined by what the OTA “thinks” it should be displaying. There are no rules in place today that says it has to be the lowest fare. So pricing discrimination already occurs in the marketplace.

        For clarity sake – I am not suggesting that airlines are the ONLY place for distribution. perish that thought. But I am saying that an airline has the right to determine which channels it uses and how. That includes the ability to say come to me only for pricing.

        And one more thing. The largest airline in Europe (by single brand number of passengers) doesn’t distribute its products by any other mechanism than “ask me and I will tell you the price”.

        Shall we go back to playing cricket now?


        • Glenn Wallace

          Imagine you go to an auto dealership, and the salesman recognizes you and shows you a red Taurus. You don’t like it, you are sure they also make a green one – you saw one in a brochure somewhere – so you go to the next dealership. They also show you a red Taurus. It turns out, both dealers called Ford and said “It is Mr O’Neill-Dunne again!” and Ford replied “show him the red Taurus! We need to sell more of them! Don’t show him the green Taurus though, we’re running short of those. Hide it around back.”

          But then imagine you go to a third dealership, this time they don’t call Ford, but ask you what you want, and they are able to show you a green Taurus – or better yet, both the red and the green one. Which place are you most likely to buy from?

          Now, it is quite possible that Ford says to show the blue Taurus too, and that the dealer forgot about it, but that’s kind of an edge case, because the dealer knows that the likelihood of greatest return is if the customer converts and buys something.

          So the problem I see in this entire story is not so much the green Taurus being hidden (though I don’t like it), but the dealer calling Ford every time you come to the dealership.

          Having a supplier controlling the selection of their product shown to each and every customer in every single channel is a pretty big deal, I can’t think of any other place where this happens today. Now I wouldn’t call it price fixing, I would call it “selection fixing”. Who is looking out for the consumers’ interest?

          Selection (both the verb and noun) is king.

          > There are no rules in place today that says it has to be the lowest fare. So pricing discrimination already occurs in the marketplace.

          Well that’s interesting, because there are a few *airlines* that DON’T want to show their lowest fares to consumers everywhere, because they believe that is a revenue dilutive strategy. It is quite possible that a small part of the reason for NDC existing is to shift the ATP higher on third party sites – though this is just speculation on my part.

          The current “rules” (from the airline) are that it needs to be a “correct” fare, validly constructed and priced. If airlines create them (low fares, high fares, inconvenient routings) who should decide which options the consumer sees?

          > Should Amazon be telling Samsung it can ONLY distribute a certain way and have pricing authority over Samsung?

          In the agency fare world, the airlines set the prices. They file the fares and everything else.
          And you’ll see that on Amazon too in some cases, manufacturer set pricing. Southwest didn’t sign any full content agreement, and their airline hasn’t fallen into ruin. In the same way that Samsung doesn’t have to sell through Amazon, airlines don’t have to participate in the GDS full content agreements if they don’t want to.

          But I imagine Amazon would fight tooth and nail to protect its right to decide the assortment of search results returned to a customer. They’d argue they know the customer better, because the customer buy brands other than Samsung from them. And the customer buys other product TYPES from Amazon too, so they know which Samsung item is a good match for the rest of their purchase.

          You didn’t answer my question about Google Instant Flight Search and NDC. The ideas seem functionally incompatible – how can you instantly search for something when you a) need to match the customer with an offer, and b) have to send a request back to the airline to decide what to show the customer?

          The computer science majors will answer, that’s easy, we can just have the supplier push out the rules and data describing the pricing scheme so Google can calculate it all locally and instantly.

          Oh wait, that’s the lousy legacy old school system we have now – distributed calculation!

          Maybe how Google will do “Instant NDC” is worthy of an article by itself, and a few questions towards Google?

          And rugby league or rugby union?

          • John Pope


            League is too up and down like basketball (of which, I played the college version), whereas, Union is more nuanced, and strategic, much like a Champions League football (soccer) match.

            To my less informed mind, definitely Union.

            “Swing low, Sweet Chariot…”

            BTW: How did you enjoy the recent Lion’s Tour? (you don’t really have to answer)

            Won’t get into the Ashes, either! 😉

        • John Pope


          After taking in all your points, processing them to the best of my ability, and wondering what pithy commentary I could possibly add to your objective observations; I have just a few important, and hopefully, relevant questions to pose to you.

          1) Which game do you enjoy most? (I’m a fan of Mixed Martial Arts, personally)
          2) Who is Chris?

          Most any other relevant queries are well above my pay grade, and I will have to defer to my more respected, examined and honorable Distinguished Gentlemen, on the floor of the House.

          I relinquish my remaining time and word count to the Distinguished Gentleman, and your nearby colleague, from the Great State of Washington.

          Thank you, Mr Speaker. 😉

          • John Pope


            Just realized I forgot to include my final query to Timothy earlier. My bad, was distracted by a rush of caffeine to the brain.

            3) Free Will or Determinism? (my apologies, if this actually rhetorical)

            “Enquiring” minds want to know.

  4. Glenn Wallace

    Putting aside whether NDC is a good idea and for whom, the technical merits of moving to a supplier-centralized calculation system for prices and selecting flights are dubious. Right now the pricing components (schedules, fares, rules, availability) are sent in many different directions, and all the different faring systems (Google/QPX, Expedia, Vayant, Everbread, the GDSs) all do their best to find the best flights. Airlines already have to rely on GDS availability caches, proxy calculators and the like to insulate their inventory systems from the shopping onslaught of the Internet. The airline systems which are now hit with a multitude of simple queries (“what is the availability of this flight”) are now going to be hit with very complex queries that will create new CPU costs for the airlines which are now being borne by the industries distributed shopping and pricing systems. It may be that reducing the indirection between fares and availability will make the calculations simpler in the long run but regardless the airline NDC shopping servers will still need to create connections, choose flights, enforce some amount of rules, select products and offers, etc…. not a simple query.

    By moving all of these calculations to a single location, the supplier, new bottlenecks will be created, and there will be no diversity of solutions because there is only a single system (for a given supplier) selecting flights and pricing itineraries. This means that consumers are not going to benefit from obscure itineraries or combinations of flights that each pricing system prides itself on finding.

    I do not see any consumer benefit in NDC, and the airlines may definitely see technical and scale challenges migrating an industry that is currently a push distributed/calculation model to a pull centralized/point to point model.

    • Timothy O'Neil-Dunne


      I must take issue with the assumption that a supplier controlled totally centralized mechanism is the only solution. All retailing systems for non travel products have a common model. The supplier takes pricing risk or the retailer does. In some cases the supplier forces the retailer to take risk. (MSRP has been accused of being price fixing in the past!) In generality we should not assume that NDC as a concept is any different from other retailing businesses. If you want a verified price ultimately it sits with the manufacturer. Travel products are very complex. Air products are becoming increasingly so. But these complexities can be easily overcome (with a different methodology akin to non-travel retailing).

      The current convoluted – fare component publish, multi-calculate, different answers – system fails to allow real time dynamic pricing in the neutral retailing channels. Further there are very few travel retailers today who can vary their pricing. That creates the tension. The airlines (particularly the LCCs) can and do dynamically change their prices. (Let’s avoid the discussion of price discrimination). Distributing those changes is a MASSIVE overhead that consumers are paying for. Moving to a retailing model including a wholesale component which the retailer adopts the risk (and subsequent rewards – positive or negative) in my view changes the paradigm and does indeed benefit the consumer – provided it is handled honestly and fairly. I urge you to read the IATA commissioned study for some examples.

      I will grant you that conventional wisdom (which has each query go to the supply host) will create bottlenecks. But we have MANY examples out there today where high performance systems can accommodate such demands. This will require retailers and GDS like companies (who should be providing neutral shopping systems) to produce better solutions. But that fact should not deter anyone.

      I can point to several examples from Everbread, ITA, and Vayant that have developed such solutions. These are independent players. Too today the GDS companies in either GDS or PSS modes have similar systems in place. Not all of them do everything but as a class they can be developed or are developed enough to provide that level of shopping capability. Meta-search has shown that shopping from an observed perspective also works (Skyscanner and Kayak et al). Thus there are different models, different solutions – all of which have their own merits available WITH and/or WITHOUT ATPCo.

      And for a reasonable/smallish fee I can tell you how to do it 😉

      But kidding aside – we must rethink the problem and the suppliers hold the key whether we like it or not. With GDS supported distribution declining; direct and independent distribution PLUS meta search maturing – we have the technology. The rather bigger problem in my view is a commercial/political one. Full Content Agreements provide a commercial structure that has the suppliers guaranteeing multiple “correct” answers. This is just insane. It worked well in former times but is no longer relevant. We need a commercial framework to provide a solution for the consumer which allows her/him to make a rational decision. Today’s solutions are in my view are working against the best interests of the consumer. (And just to be clear I am not advocating regulation).



  5. Timothy O'Neil-Dunne

    I keep asking if we can debate one topic which is the non-commercial issues of NDC. In creating a common standard for the supply side of the equation vs the current 2 tier architecture that exists in the market we create a degree of simplicity. Paul’s question is who could benefit, is a valid one. I would suggest that there are many who could take advantage of the situation. Will they?

    The reluctance to do so is driven by FUD. I am still somewhat shocked by the vehemence of those that are demonizing NDC. If we were to put the majority of the Naysayers in a room – they probably could not defined a technical reason not to deploy NDC. So it comes down to the basic issue of who will take the responsibility if things go pear shaped. Because the airlines will not do it, and there are no GDSs to take on that role in a truly open environment, fear is real to those used to being “protected”.

    If I may take a couple of points here. Adele – EDIFACT is not comprehensive enough for an unbundled airline product, even if its performance was superior. And to be clear – not everyone has access to native EDIFACT. Trying to squeeze that functionality out is a big challenge.

    So the process if flawed and the politics are bad. So what? Solving the problem is critical. And no one else seems to be coming up with a viable alternative.


  6. Adele

    There’s so much conflicting information about NDC out there, including in this post.

    Take this statement from Mr. Richer – “NDC access would presumably be good for Google, as it would essentially be able to plug yield (availability) information right into its servers, rather than using the patchwork solution that it has today.”

    He makes it sound like an application, an aggregation of availability. But later in the post he says “So under the auspices of IATA they are developing a sophisticated XML message set that will address today’s deficiencies.”

    Messages, even the EDIFACT ones currently in use by most airlines, can deliver availability to anyone who has the commercial right to the information, including Google and any other intermediary.

    The better question is “will airlines want to work with Google?” We have Mr. Richer’s definite opinion on this topic but I’m not so sure. Google is just another channel, after all, that works for some routes/regions/markets/airlines but not all. If NDC is just another method of communication between trading partners, I can’t see why Google would benefit any more than any other party.

    • John Pope

      “If NDC is just another method of communication between trading partners, I can’t see why Google would benefit any more than any other party.”

      In theory, perhaps. But in practice, the fact that Google can far more quickly and easily allocate all of the resources, and absorb the costs, necessary to subscribe to, and integrate with, all of the new technical specs will no doubt put Google into a competitively advantageous position; compared to their smaller, less resourced, competitive set who have already invested significantly in the current framework.

      Apple, Microsoft (only loosely), IBM and Exxon – their relevant market cap and financial equals – are not part of Google’s comp set in travel and air; companies who are a tiny fraction of their size and sophistication are in this sector.

      But that’s also just the nature of any fundamental, technologically managed change process, not necessarily unique to the airline or broader travel industry.

      Surely those dynamics will lead to an advantageous position for Google with their myriad of superior characteristics and capabilities – they’re no doubt loving this process, and snickering constantly under their breath, whilst other players – who will have more comparable impact and a higher percentage of available resources to lose – argue amongst themselves about all the relatively small issues and legitimacy of the process. You know, sweating the small stuff.

      All the while, it doesn’t really effect Google negatively, one way or the other, unless the airlines decide to cut them out of the loop – which we all know would and could never happen.

      Which comes back to the efficacy of IATA’s original decision not to adopt Open Travel’s schemas, as they are already the broader industry consensus and benchmark. Which also means the cost and time of adoption would be far less for many existing 3rd party distribution channels – their existing technology infrastructures would already be in place to make integration with other components and product categories that much more efficient – and the overall process to commit to and adopt the required change more frictionless.

      Nothing too serious, just likely hundreds of millions of dollars worth of unnecessary changes and technological hurdles spread across the industry by a plethora of stakeholders.

      That shouldn’t interrupt the quality of service delivered to the end customer, whatsoever. 😉

      I personally won’t hold my breath, on that one.

      When massive, fundamental change happens at this scale – in any sector – it should ideally be managed with the utmost consideration of all partners’ and stakeholders’ needs taken in to account – it doesn’t appear that those considerations were made to the nth degree. I’d submit, the focus was considerably more narrow than optimal.

      And that’s a shame for EVERYBODY involved. Sigh 🙁

      Don’t get me wrong, I’m a HUGE ADVOCATE for change – but also an idealist who would like to think it could happen in the best way possible for all involved.

      Anybody up for discussing the origins of the universe, now? To lighten things up a wee bit.

      Or, how about a Royal Baby Boy or Girl poll? My money’s on XX. 😀

    • Milind


      “Messages, even the EDIFACT ones currently in use by most airlines, can deliver availability to anyone who has the commercial right to the information, including Google and any other intermediary.

      This is the SLOW way of getting availability and that is in small pieces at a time. Each message generates network traffic and server load which the airline has to pay for.

      More sophistication is required to gain fast and wide availability information which can be priced for display.

      Massimo Morin gave some clues about this at the conference, Paul’s reference was to one example where the yield rules are “replicated” outside of the airline system to set the availability for a future date without having to ask for it.

  7. John Pope

    A fair appraisal, Paul.

    However, some of your argument’s conclusions are ultimately a result of commercial expediency, rather than genuine principled decisions that could have opened the door to further future product innovation and reduce, already high and existing, barriers to entry for present and future stakeholders.

    I think there’s no denying the fact that the truly open standards, supported by The Open Travel Alliance, would have a created an environment and conditions to ultimately maximize customer satisfaction, rather than the commercially expedient decision to adopt the closed alternative propagated by the Open Axis / ATPCO standards.

    If IATA truly had the best interests of ALL stakeholders in the value chain in mind, rather than just a few – or even one, as you speculate – distribution partners and the participating suppliers.

    I agree, customers will ultimately benefit from NDC – but that doesn’t excuse the fact that there is more likelihood that they’d benefit even more, form an even more open system. That seems pretty straightforward and incontrovertible.

    Judging by the amount of spinning tales on either side of the equation, it only makes sense that the ideal solution, is the one that is most neutral – and has no commercial investment or interest in the outcome – i.e. Open Travel, as they’ve had a long history of demonstrating in the past, already. You know, “the devil that you know” philosophy.

    All in all, I think most people from either side ideally want a better solution to the one that currently exists, it’s just a shame the powers that be couldn’t find the wisdom to adopt a solution, that with all indications based on merit, would have been THE best.

    So, to answer the question you’ve queried in your post; yes, I think it is possible that Google might be the biggest beneficiary of NDC, mostly because they have the deepest pockets, the most resources, the largest existing audience, by far, and currently the best capability of anyone to jump through all the hoops that IATA will impose on all parties – which other smaller players might not be able to justify financially, or capable of supporting technologically – not to mention the audience deficit they’re competing against.

    But, I do know of another company who are poised to take advantage of NDC, as well… a bit further off the radar currently, though.

    Sometimes, it’s not always the size of the dog in the fight, but the size of the fight in the dog.

    “Who let the dogs out? Woof… woof, woof.” 😀

  8. David Friderici

    I am actually missing answers to a few questions that come into my mind when talking about NDC and when reading this nice article.
    Based on the assumption airlines would be able to handle the traffic volumes

    1. Why should Google benefit from this at all? Google Flights was benefitting from the fact that they obviously had a lightening fast logic that no other meta search engine had. Connecting to NDC would mean moving this logic from Google or any other meta search engine to the airline. Meta search engines will only be in a role of content aggregators. No more. This means they actually would not need ITA any longer.

    2. This is a general question. What will happen with the entire fare distribution. There is actually no longer a need to distribute via ATPCO – at least not in a scenario where every airline participates on NDC. Pricing would probably only be done by the airline itself, respectively it’s interline partner.

    Just two question of actually many running through my head.

    David Friderici
    VP & Head of Product Mgmt. and Strategy – Airline Travel Services at IBS

    • Milind


      “What will happen with the entire fare distribution. There is actually no longer a need to distribute via ATPCO – at least not in a scenario where every airline participates on NDC. ”

      Until NDC takes over 100% from all other distribution channels there will be a need for current distribution mechanisms. I think we are many years away from seeing the back of ATPCO 😉 and besides it is still a US and Canada Govt requirement for published fares to be filed through ATPCO.

      • Timothy O'Neil-Dunne

        Actually that statement is not correct. There is no obligation under deregulation to file fares anywhere. Allegiant for example does not file fares and many carriers still retain fares which are not required to be distributed to all outlets via ATPCo. Remember there is no obligation to distribute fares filed through ATPCo. Indeed the vast majority of fares filed in ATPCo are deemed “private”.

        I believe that schedules must be filed in certain jurisdictions but now there are 2 players (Innovata and the newly independent OAG) the extent of what gets published and to whom becomes a matter of commercial imperative.

        That some channels have created price discrimination through MFN clauses under Full Content Agreements is a purely commercial issue not a regulatory one. Whether that is legal or not remains under investigation and/or subject to challenge.

        A key question emerges from this point: Is fare distribution via ATPCo necessary or desirable? That is a good question and should be debated, but let’s have a little honesty here. Fare distribution would be great if it was truly inclusive of the calculation. But its not. APTCo accepts no responsibility for calculation of the price of the ticket. Thus the combination of fares, rules, routings, footnotes etc etc (i.e. the price) is up to the retailing party. That it has the ability to be 100% different yet 100% commercially accurate is a core flaw in the current distribution model. If we could get a reliable calculation universally that would be much fairer for all concerned. That does not exist. Thus there are always multiple “guaranteed” prices out there. Some times just a few cents, some times hundreds of dollars.

        But what about the Googleplex, just because they calculate faster in some cases are they better? Google’s answers are NOT GUARANTEED in GFS. Indeed that makes then no better and perhaps worse than any other results system. Let me be specific: No results screen contains guaranteed answers. So online shopping is essentially a giant con. Something that is the industry’s dirty little secret and any Travel Agent worth his salt knows how to address.

        There are other structural flaws in the current system too numerous to discuss here. Frankly the only time you have a guaranteed answer is when the PNR is ticketed. Is that the guaranteed lowest fare? You be the judge.

        I am in favour of a more honest system that either lets the airline set its MSRP or the reseller sets the price and accepts the responsibility. If ATPCo’s solution determined the actual price rather than simply parroting the component information from the airline’s elements then we could have true neutral shopping. The current system confuses the consumer which is why he keeps looking and looking and looking. A core reason there is no trust in the system. And why Kayak, Skyscanner amd indeed GFS exist in the first place.

        As David indicated in the first comment what has Google trained the user to accept? Speed is better than accuracy. Go back and look at some of my former posts on this subject.

        The consumer is making his own choices and using better tools. AND she/he is a much smarter buyer than just a few years ago. Do you think you can fool all of the people all of the time?



        • Milind

          Hi Timothy,

          To clarify what I meant:

          Firstly I believe that it is still a regulation that “published” fares to/from US / Canada must be filed with ATPCO (or SITA since around 2009). Of course that excludes Private and Cat 35 fares and I don’t think you have to “publish” any fares in this way, though the legacy carriers do so because I think a lot of business processes other ticket prices are set in relation to the “published” (e.g. Y class) fare.

          Of course there are many airlines not filing private fares through ATPCO, either because they are outside of US/Canada or they are Private or Cat 35 but also there are many carriers who are filing all their fares through ATPCO because it is “convenient” to them. The distribution channels therefore have to subscribe to all the fare feeds or allow Private/Cat 35 to be uploaded direct into their systems (as the GDSs do).

          Which leads me onto the second point – I think it will take YEARS to disrupt the current fare distribution mechanisms, even with NDC whilst there are still channels that need to subscribe to the fares the most convenient and cost-effective way possible.

          But as we all know it takes a revolution to put disruption into effect, what shape that revolution may take is anyone’s guess as I think it could be a decade or more away. NDC’s timescales are still measured in years not months, In the meantime we are stuck in what was “fit for purpose” in the 70s and 80s.


          • Milind

            I also meant to mention – with fares quoted / ticketed direct with the airline (whether NDC or not) fares will be guaranteed, no ADMs.

          • Timothy O'Neil-Dunne

            Millnd I think we agree then on a number of points BUT allow me to continue to disagree with your point about the tired old 1960s designed fare systems ie the process of filing fares via ATPCo.

            Already many of the airline systems have been tuned to generate results that need not rely on ATPCo. (Internal for example, Navitaire users etc). With approx less than 50% of the world’s airline product sales going through a GDS powered channel – the need for ATPCo powered solutions has lessened. The concept of distributing WAY ahead of time your dynamic product price and then letting someone else calculate it is a far cry from distributing a price list MSRP type product from Amazon. So many forget how complex an airline product really is. A definitive reliable guaranteed price either has to be consistent across all channels or it has to be determined from a single source. If its the airlines’ product they get to decide that. Or they should be! Consumers expect it and we have a responsibility to provide that.

            There is nothing wrong with a reseller choosing to sell a wholesale plus markup (including at a loss) product. But that model is rarely in the market and usually in very specific cases such as a bundled or a consolidator ticket.

            Bottom line we should remove all restraint to how products are sold. No single party as a right to a business. There are some very capable resellers in their respective channels. But there are still a lot of “agents” who think the suppliers “owe” them a living. Harsh realities of the economics however applies both ways. Airlines should not expect to get a no charge distribution capability.

            Let’s get some reality and honesty back into travel “selling” and call it retailing. I think it will take some time but removing unnecessary complexity and embedded artificially inflated costs must be a goal.

            But wait who owns ATPCo? That would be the … airlines 😉


    • Tallman


      I agree completely with your first point. Many commentators miss that what Google/ITA has built is a new type of user experience based on speed, which requires instant calculation of availability and fares. This means that the better quality data they have *locally*, the better results they can display (and faster). NDC drives in the completely opposite direction, keeping data and calculation in the airlines’ data centres. Google using NDC would mean they lose (consistency in) speed, thus reducing the user experience.

      Anyone that doubts this can compare flight search for US domestic on (ITA has the data from their US customers) with European flights on e.g., which is more of a hack to cover a market and much less impressive.

      While it is quite possible that Google Flight could still be a big contender using NDC, it is not in line with their current technical model.


Newsletter Subscription

Please subscribe now to Tnooz’s FREE daily newsletter.

This lively package of news and information from Tnooz’s web site provides a convenient digest of what’s happening in technology that drives the global travel, tourism and hospitality market.

  • Cancel