Why the industry should wake up and smell the new distribution coffee

NB: This is a guest article by Evan Konwiser, co-founder of FlightCaster.

News flash: The travel distribution model is broken. In 2011, it just doesn’t make sense anymore. American Airlines‘s Direct-Connect is a fundamentally better way to distribute and the industry should read the writing on wall and adapt.

Suppliers pay distributors to make products more easily available, but will continue to do so only so long as that service is valuable. General Mills needs distributors because delivering cereal boxes to every supermarket is challenging and costly.

For airlines in the 20th century, outsourced distribution also made sense, as it was technologically prohibitive to easily distribute inventory to thousands of agencies worldwide.

But the world has changed (actually it changed a decade ago, but I digress…). The current business model of travel distribution is now obsolete.

global network

With the evolution of APIs, airlines are paying GDSs to do something they can just as easily do themselves. In fact, it’s the agencies that are benefiting by easy access to airline fares and bookings.

In a more rational industry, agencies might be paying airlines for the right to book on their systems. I’m not (yet) proposing that, but it’s an interesting lens from which to examine this issue.

Beyond the obsolescence of the GDS in distribution, the other big issues are price flexibility and transparency. Again, we have a business model problem. The GDSs are forcing restrictions on the airlines that both reduce the airlines’ ability to merchandise and enable hyper-competition between carriers.

So now the airlines are paying for something they don’t need that is damaging their ability to grow revenue and compete sustainably. That’s a triple whammy, and it should surprise nobody that they’re trying to change it.

Direct-Connect is a no-brainer for airlines. While we should be wary of any move that is good for one player and bad for the rest of the industry, I don’t believe that is the case here.

Open Allies For Airfare Transparency (OAFAT?) would have you believe that comparison-shopping is impossible without the GDS. That is just absurd.

The market has already proven its capability in creating price transparency through the success of meta-search engines. Kayak searches many disparate places for airfare data and then displays them easily for consumers, and OTAs and TMCs are capable of doing the same.

In fact, Kayak and Priceline appear to be doing a fine job with American fares today (Disclosure: I once worked for Kayak). Direct-Connect simply wants you to ask American Airlines directly about fares, but doesn’t care how or where you display them. As a comparison, this is not nearly the same level of restriction as we see with Southwest today, where you actually have to visit their web site.

In the Direct-Connect world, there is still a need for data aggregation to easily compare fares. However, GDSs are unwilling to put their fees at risk to become these aggregators. It’s shortsighted strategic decisions like this that will eventually lose to innovation.

Note the failure of Blockbuster, which was unwilling to sacrifice its core retail business by going through mail and online. And like Blockbuster, which is now in bankruptcy thanks to the upstart Netflix, GDSs will find themselves in trouble, getting beat most likely by ITA, Farelogix, and other technology players. The companies that innovate even when it compromises their core model are the ones that survive disruption, not the companies that resist and form defensive alliances.

Finally, there has been concern about the implementation of personalized fares that are enabled by Direct-Connect. Some would cite this is as an anti-consumer tactic. Nothing could be further from the truth.

Personalized fares offer just as many ways to offer ancillary freebies or discounts as they do to beef up fares for the price insensitive. Airlines are still subject to the laws of supply and demand.

The invention of fare classes enabled airlines to differentiate business from leisure fares. Overall, that was a great innovation for the industry. It enabled the cheap-enough leisure fares to engage the public in air travel, and asked those willing to pay more for better service or fewer restrictions to do so.

The airlines understand that even business travel has a supply and demand curve – trust me, they’ve felt it pretty hard these past few years. We in the travel industry should embrace fare personalization to deliver better products to more people at more relevant prices, not fight it because we’re wary about disruption.

With Americans Airline’s initiative, the status quo has been disrupted. But putting up walls to innovation, creating alliances to lobby deaf ears, protecting a legacy business model at all costs — these are recipes for disaster.

You may hold on a little bit longer, but someone is going to come along and destroy you, and for that onslaught I’d look westward, perhaps towards Mountain View, California (home of Google) and the one big player who has yet to comment on this issue.

NB: This is a guest article by Evan Konwiser, co-founder of FlightCaster. Follow him on Twitter or read his blog.

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. GDS User

    how are codeshares, cross-carrier ticket validation, and baggage agreements going to operate in this new system? I just can see how direct connect replaces the global functionality the GDSs provide….

    • Timothy O'Neil-Dunne

      Can you clarify – you can or cannot see how DCs replace?

      I am assuming that you believe that it cannot. Each airline is free to allow any carrier to interline and then also to administer their various agreements. These are base level requirements from any system and exist in airlines, legacy GDSs and in the mis-named Direct Connect systems. Indeed some of the newer systems have taken a very innovative approach to these issues.

      However in the case of the legacy GDSs not all things are administered equally. Further there are issues regarding the ability of a single airline’s willingness to take responsibility for an agreement even one that is signed and accounted for. EG in the UK a regional airline in many cases will refuse to transfer bags at a major London airport if there is more than one downstream sector involved. There are many practical issues that it is assumed the GDS handles when in fact it is the airline who takes that responsibility.

      I hope this answers your question


  2. Bruce Sweigert

    I understand the argument on airlines saving cost and wanting to distribute their product the way they see fit, but from the customer’s perspective, is this the right direction?

    I would argue that the trend toward unbundled fares has not been truly in the customer’s best interest. In the past, it was fairly easy to compare air fare offerings on a like basis; these days you never seem to know what the all-in fare is until you actually pay, and sometimes not even until you get to the airport to check in.

    What the consumer enjoyed with the old model was the fact that they could go to just about any channel and get a pretty consistent view of fares – whether it was through an agent, OTA, metasearch or airline.com. The customer was free to choose the channel that made the most sense from both cost and service perspectives.

    With the content deals in place, the GDSs were enforcing a state of standardization of content and fares access across the industry that worked well for the customer. Perhaps you could say it worked too well and there was too much pricing transparency that resulted in an over-competitive market and resultant falling yields that he airline industry has experienced over the past 15 years.

    Now, with the growth of ancillary offerings and fare unbundling, the shopping process all of a sudden got a lot more complex. If that was not enough, with the arrival of the recent large carrier feuds with the OTAs and GDSs, and these carriers withdrawing content, or preferenceing one or more competitors, further compounding the complexity for the consumer. Not only have fares become more difficult to compare, customers now have to do these comparisons across multiple channels.

    With this new state of chaos that the travel distribution market is in now, the hope seems to be that search will come to the rescue of the consumer and bring standardized comparisons back to the travel purchase process. Will this be the case, or in fact will it become even more daunting as airlines are empowered to create an even more complex list of offerings, and differentiate their offerings by channel, and perhaps even by customer. Introducing pricing opacity back into the market may actually be the ultimate goal of the airlines.

    It comes down to the question – are more complex, more opaque offerings across multiple channels better for the customer, or are simple, transparent, standardized fares better? Ultimately, will the abundance of new offerings and savings from GDS bypass, supposedly passed on the customer, put the customer in a better position or not?

    • Timothy O'Neil-Dunne

      Bruce, you make a number of very good points here. Not least of which is that the complexity has to be addressed. But there are some points that perhaps you have not considered.

      1. The consumer is already smart – he already shops across channels and in different environments. This is not new behaviour. For the US consumer who has benefited from the economies of scale this is going to be harder in the short term. Longer term though they already do search in different places so it can be expected that they will continue that trend

      2. That the legacy GDSs have powered simple search – albeit in an adequate fashion for so long – is no longer a guarantee that they can continue to provide that solution in a more complex world. Indeed they have proved themselves inadequate both in the long term (IE – ITA showed there is a better way across neutral product – and the near term by failing to provide suitable products for either the B2C or the B2B user) and while having had the writing on the wall about ancillaries laid out for them for more than 5 years – they have ignored the demands of their paymasters and their user community – not to mention the customer at large.

      Their response to the situation was not to innovate but to fight back with lawsuits and monopolistic behaviour.

      The consumer then has been disadvantaged for quite some time without realizing it.


      • Bruce Sweigert


        Yes the customer is smart. Right now, pretty much all I have to do is go to Kayak, then also check the few carriers I know that don’t participate, double check directly on a few the airline.com sites then pretty much that’s it I’m confident my shopping is done. Actually easier than it sounds because I have my preferred brands, but I don’t live in the USA that’s why I have more steps.

        I agree with your comment on ancillaries. Here is a trend that its seems obvious that customers don’t like. So why is it that we need to give up a neutral environment to empower something that we don’t want?

        But my issue is how much smarter do we have to keep getting just to keep up. On situation I’m worried about is a future where, after consulting multiple channels, then I have to log out of my browser, clear all the history, cache and cookies, turn on the private VPN and then do the same search all over again. I’m sure there may be many price in-elastic customers out there that are perfectly happy with the fact that they paid a higher price for exactly the same product and service, but how many others will now find themselves spending even more time to find the best deal?

        • Timothy O'Neil-Dunne


          I think my point was that you don’t need to give up anything. The assumption in “giving up” something was that you had a truly “neutral” environment anyway.

          In the non-US part of the World – the GDS is understood not to be the ubiquitous form of neutral supply chain access any more than say Amazon is the only source for books. In the USA the assumption has been that with the exception of Southwest – the legacy GDS gave you everything and therefore anything that was GDS powered was good enough. This is not true outside of the USA and now not true in the USA – except only now is that becoming apparent.

          As an example just compare http://www.Skyscanner.net with say http://www.Expedia.co.uk. You get the very clear picture between the two. It doesn’t mean that Expedia UK is not valuable – just don’t expect it to deliver everything because clearly it doesn’t. Thus Expedia UK is rather like Amazon – you can find most things there but if you want the best at the lowest price you have to look harder elsewhere.

          In the USA market the Airlines can make a claim – no lower fare can be found anywhere other than the airline.com site. In the rest of the world with greater opportunity for off tariff pricing – then search on different sites and different channels to get better pricing or service packages. I can walk into just about any agency in say Germany and get a cheaper price than on most airline.com websites. So the agency is doing a good job for me and providing me with a service and I can expect that. (Plus if I am German – I prefer dealing face to face).

          To your final point – the need for clearing the cookies? well to be frank everyone is being gamed on this and so you should clear your cookies anyway. But people don’t. New browser versions e.g. IE8 and IE9 etc let you start a fresh clean session with a single mouse-click combo. The emergence of newer privacy protection services in the browsers make that possible. I believe in the consumer being smart enough to know when he is being gamed. He will search harder. The truth is out there …. somewhere… just not necessarily legacy GDS powered. And the consumer knows what to do – AND she/he does it.


  3. Sebastian

    Hi all

    I have been working in the distribution department of a large Europen airline for some years, doing negotiations of GDS fees. I also know the GDS business from inside. I would just like to make some comments to this great and helpful discussion.

    1. The number of cases where even big airlines achieved significant net GDS cost savings is neglectable. What GDSs say in press releases and the reality are too often two very different things.

    2. The value of each respective GDSs achievement in building a relationship with TMCs or OTAs is zero. Every travel agent uses a GDS. For the airline – despite fees – a booking via Amadeus is the same as a booking via Sabre is the same as a booking via Galileo. GDSs argue their inflated fees with an increasing market share or increasing number of users. But that’s completely irrelevant from an airline’s point of view.

    3. The main user of a GDS is the travel agent. Most functions in a GDS are to make the travel agent’s life easier. One of the big GDS told us that more or less half of the functionality around air reservation (not taking rail etc. into account) is aimed at travel agents. But travel agents don’t pay for it.

    4. An example: A GDS booking costs $7. 50% of the technology paid for is travel agent technology. Remains $3.50. $2 are paid out to the travel agent as an incentive (in some markets it’s much, much more). Remains $1.50. Every airline I know is willing to pay $1.50 for a GDS reservation. It’s more than what airlines pay in direct sales, but it’s worth the added value the GDS provides. If the travel agent would pay 3.50$ (its $3.50!!!) and the useless incentives would be eliminated, the margin of the GDSs would remain on the same (unbelievably high) level of about 30% or more. Travel agents would (finally and at least 10 years late) need to make up their minds about the value they provide and reflect that in what they charge their customers.

    I wanted to add these comments to the discussion because I feel they might get lost on the way 😉


    • william

      Good point …
      Everybody in the chain is taking its dime (like in all distribution channels)!
      Question is then do people want to give 3.5 dollars to a human that will help you do your travel booking (and maintain jobs) or do you prefer doing everything on the web or via airlines call centers?
      At the end, anyway, nobody will give the traveler that money back.

      • Sebastian

        Quote william: “At the end, anyway, nobody will give the traveler that money back.”

        Correct, and in economic terms that would be the price adjustment needed in an industry dominated by a monopolistic and inefficient business model. Let’s face it.

        • william


          Looking at the numbers, I do not see a monopolistic business model here …

          GDS are still the major providers of booking for airlines in the USA, but I do not think we can talk of global monopoly. In NORAM, 63% of air sales where mde by GDS and 47% in EMEA (phocuswright report for 2006-2008). in hotel it is around 10% and even less for car. BUT THEY HAVE THE CHOICE TO USE OTHER DISTRIBUTION CHANNEL AND TO CONVINCE THEIR USERS TO USE THEM.

          Again, I would like to understand how the model proposed by the airlines could offer a fair and non-monopolistic business model.

          American Airlines is also pushing a monopolistic business model. It acts like a luxury brand, selecting who can push its fare and who can not. Looking also at how their fare should be pushed and presented off the shelves.

          What is proposed is to shift the monopoly from one group, to another. I understand then that some people do want to resist … Worse, the traveler is not going to benefit the change …


  4. Timothy O'Neil-Dunne

    Wait so the world isn’t flat any more 😉

    These are all valid points. I think we agree that change is coming like it or not. And open better mean open or else this effort will not have advanced anything.


  5. Tim

    I find this subject fascinating. I’m nearly 50 years old. And even I know how to open my browser, open 3-4 tabs for my preferred airlines’ web sites, enter the dates and destinations, and in the blink of an eye my search is complete. Now I review the offers, the departure times, etc, and make a decision. It could not be any easier. And I’m OLD!

  6. william

    It will be difficult for me to still have something valuable to add 😉

    1) for me “Direct Connect = Private API” > Nobody knows today if a “direct connect” supplier will not provide “advanced” API to some, and a generic one to all others (remember Microsoft). GDS are offering their API and content to all.

    2) GDS air content is synchronized between GDS through OAG. Direct connect are alone in the cloud. Imagine a DDos Attack and nobody can buy tickets for hours … Remember the ash cloud, CRS booking were blocked since nobody was able to product when planes could fly over Europe. Look at the numbers of transactions managed by Sabre, Amadeus and Travelport … Who can ensure airlines will invest enough to sustain the load?

    3) NORAM and EU and China and BRIC are not the same market with the same regulations. In china, the ecosystem is moving towards a GDS approach …

    4) Who will develop front office solutions for Travel agents all over the world based on direct connects? Do you imagine how many people will have to use new tools, new processes. Who will train them, the Airlines?

    5) Direct connect will evolve a lot, GDS are more stable … If a company want to change its API to earn more money, then, guess what? Connect or die.

    GDS are needed for Travel Agents and global corporate customers.
    Direct connect is to be leveraged for customer facing online leisure oriented suppliers (like OTA / OBT) and by airlines building their own web site and mobile solutions. This is a no brainer. It is mainly a commercial dispute.

    It is also the birth of Travel 2.0. Each company will baked its business within an API to try to benefit from innovations all over the world (the world is flat). As Sam Ramji (VP Strategy, Apigee) said, it is not just about selling in, it is now about selling through (API). You will shift the power from GDSs to thousand of companies, developpers, building great web/mobile/TV apps. But at the end, this will only cover part of the travel ecosystem need.

    Connecting to different API is not just as simple as connecting a USB key in your laptop. Semantic mismatch, technical issues will grow over time making integration more and more difficult (that’s why EAI was invented). And since the travel ecosystem always refused to build/accept/use technical exchange formats and standards, then, we are back to anarchy. Or as Phocuswright CEO said: chaos.

    One size does not fit all …

    • Daniele Beccari

      Agree (nice reference to Apigee) but one quick note:

      “the travel ecosystem always refused to build/accept/use technical exchange formats and standards”

      Well, the airlines came together to create AIRIMP in order to build all interconnectable systems, and it’s been working beautifully well for decades.

      But that’s not enough, as you point out, APIs are a complex thing. 1 API connection is simple, 2 is complex, 3 is a nightmare. Imagine 400.

      • Timothy O'Neil-Dunne

        Seems you two share things in common other than language!

        I wish we could stop using the term Direct Connect – but its about as much of a misnomer as GDS and LTDs.

        Today the GDS market supports an astonishing array of different connection types. The data standard – AIRIMP manual is a mind boggling series of exception codes. But that is the way the world is. And yes its complex. I realize that both of you have direct knowledge of that complexity.

        But I would like to point out some flaws in your argument. The APIs are actually not dissimilar. The core root of many of the APIs out there are for the Open Travel type and W3C compliant. Each GDS has its own. That makes 4 to start off with. Even if Travelport succeeds in delivering a common API to its 3 GDSs and Sabre to its two that is still 6 APIs. So talk of a single common API is not a reality.

        The belief that there is a single source for all content is also not realistic. There are over 100 LCCs who in some shape or other provide APIs to their hosts. Also the legacy GDSs do not agree. Anyone can demonstrate the differences easily – the calculation engines are just – well – different. According to the rules – they are all right as no exact reference point exists. Even the airlines make mistakes and miscalculate as I demonstrated on my own blog this week.

        Thus we are left with the reality whether its 4 or 20 – there will always be a lot of them. If your mission is to provide ubiquity of content then you are obliged to deliver multiple connectivity elements. I call this the problem of being a little bit pregnant.

        Addressing William’s worry about scalability and multi speed APIs. We already have that too. However there are already fine and robust implementations of high grade multi access systems in place today processing large quantities of transactions using literally dozens of APIs at the same time. Each day thousands of bookings are made via these systems. Indeed without them the meta search engines would have brought down at least one GDS.

        Where I will agree with you – is that this work is not trivial at all. I have been working on systems that connect into the core GDS infrastructure for more than 25 years and it is not easy. Its incredibly complex. But the driving that to the single conclusion that the GDSs are the only way forward is a the same argument that Henry Ford made with the Model T and about just as relevant in today’s heterogeneous world.

        Consumers want choice and they want the data their way. The fact is that the legacy GDSs cannot provide quality and fast search is why ITA and its like sisters exist.

        The agents of today who are confronted with these options can provide a better solution in a diverse market but they dont scale well because their tools are not up to the task. Better tools are now available that allow an agent to search faster and with greater choice than the GDSs have ever been able to provide.

        Plus Ca Change….


        • william

          Well, well…

          Creative destruction is a well know concept in economy.

        • william

          Well, well…

          Creative destruction is a well known concept in economy. Of course GDS could disappear and the whole ecosystem around them also. And of course something new will emerge. Or not.

          For complex booking and ticketing, aka multi-leg segments, I still need to see how that could work with multi-direct connects.

          Direct connect will also mean, direct payment to the airline, since most of content hubs (like ITA) do not offer it. So, If I buy a ticket with a direct connect, then, only the airline could do the change, refund, etc. If you book from different airlines then things are different. Cancel and refund process are also really complex when you deep dive into details. Sometimes, you need to wait 24 hours for refund … Will you stay in front of your browser waiting 😉

          My main question is what is SO GREAT with what American Airlines or JetBlue are offering today? Ancillary fees? What makes a traveler more happy with the revolution they propose? A more expensive ticket, a worse travel experience (3 in a row on the right, three on the left), oversold tickets, etc.? I do not see the innovation here, just a tentative to create a walled garden.

          And when iTravel from Apple, RIM travel, Google Travel, Facebook Travel will all require to use a particular phone, a particular platform, tool or web site, you can imagine how easy it will be … All those guys will try to earn money also … The question is how much apple will ask: more or less than today with GDS? More or less than today with the newspaper?

          And then you will have less and less suppliers willing to pay for offering the whole content. They will prefer to integrate with the more profitable airlines, hotel chains, etc. Fragmentation again and Internet for all.

          Look at hotels and how much they pay to OTA, with their direct connect. Do you think the situation is better?

          If the whole story is about 3 or 4 airlines trying to earn more money, why not negotiate with OTA and GDS both technology and pricing? Why wanting to kill revenue streams and a whole profession globally?

          On line adoption varies enormously from country to country. The ash cloud proves that in case of a crisis, airlines are not able to cope with the massive load of “customers” on phones, facebook, twitter. Travel agent are there to humanize (not sure it exists in english) the process. Sometimes you need to talk to a human. Travel agents are trained and use GDSc today.

          Airlines, like Hotel, should accept and benefit from different distribution channels with different business models. When a business model is no more adapted, then it needs to be modified/re-negotiated.

          Of course, you can replace GDS with Google/Kayak/Expedia search, using ITA platform, connected to all airlines using direct connect. But I’m still wondering how you will modify the web site to show all new fees and how you will compare them between companies. What will be the booking rate is also interesting! Of course, you will sell lots of coach seats … for the others …

          And to conclude: unfortunately other airlines from other countries are using both GDS and direct connect and are growing fast. The danger with creative destruction is that you never know how much the disruption you created could harm your business and your ecosystem in such a flat world.

  7. Jonathan Alford

    Excellent article, Evan, and everyone here has good points. Fascinating space.

    Can I add another case to Evan’s Blockbuster example? In a prior life I worked at Circuit City, which used to dominate consumer electronics, built on its salesperson model where suppliers paid incentives to the company, which in turn paid them through as SPIFFS to salespeople. Consumers at that time needed to be educated more on the product features and technology, espec TV’s and computers.

    In 1997 or so, people actually thought Best Buy might collapse – it was trading only at about $7 on $6B-$7B revenue, and Circuit City was trading around $50 and growing fast on about $8B revenue.

    In only about 5 years, BB was generating $25B and Circuit City took a while longer to die, but its fate was basically sealed. It tried desperately to change its salesperson model b/c consumers were becoming more educated, didn’t need the salespeople, and wanted to self-select their product. BB designed its model to that.

    CC’s salespeople did everything they could to protect their model, but ultimately the whole $10B company died b/c it didn’t reinvent itself in time to what consumers wanted, which was to make their own choices and prioritize their own features without the intermediary salesperson obfuscating them.

    Not a direct parallel of course, but history’s always interesting…

  8. rick

    “they can just as easily do themselves.”

    That is a big leap. There is an assumption that gds systems are charging fees for nothing. What they give is a wall between crazy poorly written implementations and traffic and themselves, not to mention maintaining the infrastructure for this. There is an indirect cost savings in the central gds. Now each carrier would take on this portion and watch the pendulum swing back: “hey guys, we’re all blowing cash doing the same thing, let’s consolidate and form a gds!”

    Carriers worry about look to book an awful lot, just wait till it all goes direct!

  9. Nadav Gur

    Evan – first congrats on writing a piece that seems to be getting everyone who’s smart and / or loud responding…

    While I agree with many of your premises (e.g. “from a technology perspective there is a better, cheaper way to build a search / shop / book engine than a GDS”), we should remember that we’re talking about a distribution channel, not just a technology. In traditional retail, the distribution machine keeps 30-70% of the consumer price. Why? Not just because “there is a cost involved” but mainly because it’s the retailers who, for one reason or another, manage to draw the customers. What the GDS have been doing their best to protect is not the technological superiority, but the relationships with the retailers who have the best customers – the TMCs and to a lesser degree other online and offline agents. As long as these channels have some of the best customers, there is value in paying them, and paying the GDS to get to them, and technology has little to do with that. Certain airlines can say “you know what – I don’t care, I’m competitive enough as is and I’ll get enough customers without these channels” (Southwest). American Airlines is trying to see if it can go half-way – keep those channels but supply them directly and not through “distributors” – the retail equivalent of a GDS.

    As long as there are customers who are interested in buying through such a channel, whether it’s because it provides better service (e.g. a TMC – like buying insurance through an agent) or bundles the purchase with other products (e.g. an OTA – like buying ingredients for a cake at a supermarket), these channels will command a margin, and share it with whoever enables them to get the merchandise. OTAs should be able to relatively easily switch to an alternative method of supply (e.g. Priceline adopting Direct Connect). Small TMCs and other agencies – not so much.

    And the GDS? They will have to adapt and change – or downsize…

  10. Pasqualina

    Excellent post.

    AA has started a trend and many others will l believe will follow suit, its just a question of time.

    @Bruce GDS need to start thinking and quickly to come up with strategy to stop other airlines leaving.

  11. Daniele Beccari

    I disagree with 80% of the above commentary.

    The 20% where I agree is that GDS distribution today is mostly paid by airlines, who are right in trying every idea to lower distribution costs. This has happened for the last 15 years. Every few years there is a new threat to the GDS model, and the fees go down in turn.

    I see one big difference this time, and that’s the consolidation of the US airline industry.

    When there are only 4 big airlines left, the costs for any site to implement 4 DCs is in the range of possible. (As a paradox, that’s when search engines like ITA start losing interest. The GDSs have much more than just search, so they can still add value to someone, as far as they can find a “more rational” business model.)

    It will still be a total waste of money and time if every site has to independently develop and most importantly maintain overtime 4 DCs.

    But it’s already less complex than doing 10 DCs. And it would not be airline’s money being wasted.

    • Timothy O'Neil-Dunne


      If I may I will challenge your issue on fees. I am afraid you are incorrect in your assertion about the GDS fees. The fixed fees have been “capped” but the GDSs learned a long time ago that unbundling the GDS fee was highly profitable for them. Having audited several airlines GDS bills I can assure you they NEVER go down.

      GDS fees have risen every year – even long after caps were introduced. The rise is higher than the increase in traffic and at a time when the GDS share continues to fall. Just look at Travelport’s numbers. Only when the company lost a monopoly market did their revenues fall – however after the adjustment – the numbers started climbing again.

      Yet every year more and more cash is paid out in incentive fees.

      If you calculate that today more than 50% of all airlines flown passengers of all types are now on an alliance flight then the number of direct connects are only 3 plus a few disparate direct connects. Depending on whose numbers you use that figure is likely to be now in excess of 60% with a further 5 percentage points already in the pipeline. With LCCs non-GDs flights accounting for about 23-25% that only leaves less than 20% of unaligned flights available to be shown. Does that make sense to have the overhead of the GDS?

      The model has been broken for a very long time – except the artificial life support that maintains it feeds a lot of mouths.

      The lack of broad search actually raises the costs to the consumer. Remember that the GDSs refused to provide search to the airlines which enabled ITA to find its niche.

      I agree with you if you assume that legacy solutions for connectivity to the Airlines model has to be developed. However there are MUCH cheaper solutions that are easily deployed for the intermediary – On or Offline – out there. If anyone is interested – then I would be more than happy to demonstrate several different solutions – that are currently in production.


      • Daniele Beccari

        Tim –

        I can disagree on several statements, but to remain focused here are at least two public examples where a GDS had lowered their fees. And we are not supposed to know what happens in private negotiations.

        See the Amadeus value based approach details below. Note from the definition below that AA domestic bookings (hence the absolute majority) would fall under the “standard value” case with -5%.

        More recently we have read about Sabre trading “full content deals in exchange for lower booking fees” with discounted fees of 12.5%. Again just public statements.

        I would not be surprised that at the same time GDSs have introduced other services or fees and the overall bill might have gone up.

        The 2004 pricing will apply the following percentage changes, broken down by region, compared to Amadeus 2003 pricing.

        % change in standard value booking fee
        Europe -5%
        North America -5%
        Middle East -5%
        Latin America no change
        Asia Pacific no change

        %change in premium value booking fee
        Europe +5%
        North America +5%
        Middle East +5%
        Latin America +4%
        Asia Pacific +4%

        Standard value: domestic and intra-continental reservations made within an airlines home/prime market.
        Premium value: inter-continental bookings, domestic and intra-continental reservations made outside an airlines home/prime market.

        • Timothy O'Neil-Dunne


          Using only pubic current information, let’s agree that we are speaking of the core GDS airline for distribution revenue. The issue is whether the airline pays more in total for distribution (vs for IT Services) and the GDS gets more gross revenue and then passes more onto the market further tying the intermediaries to the GDS.

          In that case the numbers are clear – the GDS have increased their margins on the GDS business at a time when – as you correctly state – the specific segment fees have been frozen or even reduced.

          I would challenge the notion that the GDSs have seen a reduction in gross revenues from airlines for distribution services. All this at a time when the global GDS market share has fallen. The numbers dont support that. Further Amadeus has seen its gross margins across all the lines of business rise.

          As we have seen from the AA/Travelport battle – both Travelport and Sabre unilaterally raised rates to that particular airline. Travelport internationally and Sabre across the board. (Now as we know a court ordered Sabre and AA to return to an as you were state).

          If we are talking about free, fair and transparent then perhaps the GDSs should open their kimonos so we can all see what they take in and what they pay out.

          Then we could all know for sure who pays for what. Perhaps the GDSs should get their wish and be treated as a utility and then regulated in disclosure as such. No that is not a pleasant thought.


  12. Joe Buhler

    Excellent commentary. Especially like the line “in a more rational industry” which hits the nail on the head. Looking back, ever since the internet first came on the scene and started to disrupt travel, the initial response of the legacy and status quo defenders was to oppose change and innovation. It is no surprise to me that they are fighting this latest development tooth and nail as well. The winners will be those who recognize the challenge, tack and ride the wave of innovation.

  13. Jonathan Spitz

    Evan, great points.

    Question: Does this all lead to a world where consumers are left with just one source for true aggregated flight data for comparison: Google/ITA?

    • Evan

      I don’t think so. I think if ITA enters this business (and I would if I were them), they will be a formidable competitor, but I don’t see why Farelogix and others can’t do it (including perhaps a new start-up or two). I also think the current GDSs can do it, and do it well, since a lot of content will still come from them — especially hotels and smaller airlines.

      The one thing I’d be nervous about re Google/ITA: They can give it away for free (or very cheaply). Nobody else can afford to do that. That, in my mind, is the problem for the industry.

      • Jonathan Spitz

        Isn’t that the point? If we strip any profit from the GDS channel and squeeze all the others, there’s only one viable model left.

        Play out the chess game…how does Farelogix do head to head against Google?

  14. Bruce Rosard

    Great post Evan. I think you’re right on with a lot of your points – it is (past)time for the GDSs to make the necessary changes that are coming.
    Here’s a few questions:
    What would you do if you were a GDS today? Maybe @saykay can answer that one.
    How do the travel agents and other distributors that won’t be on direct connect (because they’re not big enough) get access to airline fares from around the globe?

    Great to see a Class of 35 Alumni out there with a strong and (I think) accurate point of view!


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