on 14 December 2009 
(No Ratings Yet)
Recently we have seen a spate of travel startups describing themselves as a GDS – Global Distribution System.
I am intrigued by TourAbout, a social GDS. Then there is Kumutu, an adventure and activity tour GDS.
So are they really a GDS? And what is a GDS anyway? Is this just semantics or are the legacy (they won’t like that) GDSs being taken on by new entrants?
Lets have a go to define what a GDS is. Perhaps this is an impossible task, like defining pornography, where the US Supreme court said it was tricky to define but you know it when you see it.
One thing for sure, a retail travel website with an affiliate programme is not a GDS. Agreed?
OK – onto the 10 checkpoints
- No content editing – if the supplier puts in a really bad description of their product it is down to them to fix. The central GDS does not edit the content.
- End point contracts – the travel agent commercially contracts with the hotel chain / airline / tour supplier. The GDS doesn’t even need to know the detail of that contract, they are just the data conduit. (This contrasts to where, for some web systems, the supplier contracts with the website company, the affiliate / agent contracts with the website company – hence they are centralised contracts)
- Purpose of the GDS is to communicate live product, dates, price & availability data – this can lead to a transaction (booking) via the GDS or something else (e.g. a PPC click, if the company is a media model company). Many traditional GDS systems revolve around the objective being a transaction rather than marketing but I digress.
- GDS is commercially agnostic – they shouldn’t care which products are returned first to a product search query. If a search order is required (it will be) then it should be fair and equal to all suppliers based on some form of random order
- Able to apply own business logic to front end – if an agent / media company wants to use a GDS as a source of products to run a media model they can. Or an agent can run a transactional model. The GDS shouldn’t care what kind of front end model is being applied (again, the traditional GDS model is based on transactions, but that has to change at some time)
- Open to all to take part – tricky one – but say a hotel chain wants to join a GDS and a similar hotel chain is already on there. There should be no objections to the new hotel chain joining. In this sense a GDS is a true aggregator not a curator of product.
- Trusted intermediary – the GDS acts as the central, trusted, tracking mechanism for all bookings between supplier and agent. The GDS acts neither for one or the other.
- Global – the same product can be represented in multiple languages sharing the same prices / availability. If not global it is just a DS right?
- GDS charges based on the technology burden not on the value of the transaction – doesn’t matter where the decimal point is on a piece of data. Tend to charge based on how much traffic / load placed on system.
- Able to use GDS for own channel pricing – say a supplier wants to give a specific price to a particular agent. The GDS should permit end point to end point specific price contracts. Remember the GDS is just the data conduit….
Did I miss any out?
Related posts:
I’m sure you’ll get many comments on this post, but if you are defining them as they exist now, then the point “GDS is commercially agnostic” while true, is only a partly true.
The current GDS’ can, and do, bias display of supply on their screens, through ads or paid placement, very similar to sponsored searches on Google.
There will always be suppliers who are willing to pay to get their product noticed, so I think any future GDS-like entity will offer the same functionality. Why wouldn’t they, especially if suppliers are willing to pay?
How about regulatory compliance – or “the ability to accommodate regulatory compliance” – such as PCI and Security Policy Enforcement?
Hi Valyn
The commercial order that GDS systems send back data is now less important than it used to be.
Historically agents used to access GDS system via a command line – or via software that re-presented the GDS data as it comes out of the GDS. Now agents (and consumers) are much more likely to use software / services that have an additional logic layer between user and GDS data – hence the actual data order returned by GDS is less significant than previously it once was. (filtering is important, order less so)
@Bonnie – A tricky one if the GDS is to get involved in ensuring compliance from both sides of the marketplace. Ummm….
Alex, you are of course right about green screen presentation of the current GDS’ but given the popularity of sponsors and keyword buys on Google, one has to assume suppliers on any GDS-like site in the future will want the opportunity to effect order presentation. True agnosticism in display may not make business sense in all segments.
Don’t forget one point: this big systems are in an juridic low ecosystem than can be stressfull for them.
I meen, they need to deal with European commission for no competitive exclusive advantage and things like that. I guess in US it could be the same.
And they spend lot of money in lobbying and other communication stuff to explain their positions.
Small tourism operator are under the radar of this stuff
Best
Claude
And of course you forgot the most important one – traditional GDSs make their money by charging the suppliers a fixed and often quite high “segment fee”. I don’t think it’s related (in most cases) to the amount of traffic – it’s related to the number of bookings and the type of booking (e.g. Amadeus “value pricing”).
There are other mini-GDSs that do not in fact charge the suppliers anything but take their revenue from the intermediary that is using their feed. These newer GDSs (such as CarTrawler) allow zero cost distribution in that case.
Imagine if Google were a GDS…
Hi Bobby
Thanks for your comment
Surely number of segments directly correlates to system load? Value pricer certainly does (as that requires many price combination checks which is a high load intensive calculation)
I was really setting out theoretical business requirements for a GDS rather than describing how any individual company currently operates!
Cheers. Alex
I think you are placing far too much importance on the term GDS. The real opportunity for emerging segments like tour & activity, things to do, and vacation rental is micro-distribution. Global distribution systems are not required in the scenario where standard connectivity and one-to-one relationship building is no more or less complicated or difficult then connecting to a single pipe. In the case of the airline GDSs, the value was the dedicated connectivity to the CRS. In an ideal (i.e. technologically current) World, the CRS would have a standard XML connector that allowed any reseller, site, or partner to digest supplier content and do what they want with it based on the contract terms set out by the supplier. This allows partner resellers to connect to suppliers directly without having to code to unique connectors each time. As a result, local or niche aggregators can connect with any number of suppliers and feed both content and real-time availability and pricing without having the added cost of a GDS. The GDS doesn’t even enter into the discussion because it is irrelevant in this scenario. In the case of air and hotel, there is already too much dependence on the distribution capabilities of the GDS, for emerging sectors, there is no such distribution. The job then becomes for system developers who are creating tour operator CRSs to build XML connectors to a common standard so that future integration is simple for the reseller.
My analogy is this:
GDS = AOL/Compuserve
XML = Web 2.0
This is a very interesting discussion and building GDS based businesses has been what I have been doing for the last 10 years. I believe there are two essential components that represent the underlying concept of a GDS marketplace:
(1) Global & Comprehensive
(2) Ability to transact with suppliers through a standardized API (ideally XML)
If these two factors are present, then you have in my opinion what constitutes a GDS. To me XML is plumbing that provides an easier conduit for connecting.
Nice post Alex and good discussion here.
I’d add that the ‘Global’ part should support display, currency and availability by ‘point of sale’ (from a product perspective Tourabout supports 8 of the 10 points raised.)
On the ‘Social GDS’, given the lack of precedent, in some respects we planted our tongues firmly in cheek when we coined that phrase. But we wanted to define the concept of social distribution in a similar way.
Its the idea that Travel companies can create ’social assets’ and have them distributed to many locations to be consumed.
It’s distribution and its global…maybe we’re burning a sacred cow
I’d like to hear from travel agent on this topic, as one of the values of a GDS to travel sellers (and they still have a role in this industry, even in the emerging segments) is the aggregation of content so the seller can efficiently find options for their customer (this holds true for the B2C model too).
Lots of companies use the term GDS for their new products but lots of content in lots of different new ‘GDS’ buckets does not equal aggregated content, or ease of search.
Forrester put out a survey earlier this year indicating a very high level of discontent with online travel shopping. Calling a product a ‘GDS’ but offering only niche or limited content seems like a sure-fire way to annoy travel sellers and travel buyers.
Well, I can’t speak for the techy angle all I can say is what a GDS (aka CRS) means to me, as an agent. It is nothing fancy. It is to me what a spanner is to a mechanic. Simple, straightforward and does the job. No bells and whistles, nothing too clever. We all use the code entries still – there may be the odd bit that is scripted or has a fancy window, but (probably) 99% of agents have and always will work using the codes. They are faster (for us) and much more efficient. What I expect, is that if I ask for A to B it gives me A to B on one screen, invariably in time order and as I can see the booking classes, I can instantly see what’s what – or more specifically, what’s how much. True, you need to know what it all means but then so does the mechanic – having the spanner does not fix the machine, you have to know which nut to turn.
Indeed, since the late 80’s (or whenever it was the thing was invented)hardly anything has changed – mainly because it does not need to. It shows connections (across airlines) and (most importantly) does the fare construction for you. Many agents (like me) are system loyal. Personally, I am a loyal Sabre man – until the boss foisted ruddy Amadeus on me). Personally (and I do stress a personal opinion) I find Amadeus time consuming and unwieldy. I like Galileo yet found Worldspan a ruddy nightmare – but as I say that is very much a personal thing and there are as many agents who will have different points of view. I don’t mind (!) if the GDS systems want to produce XML feeds, horse feeds or any other bits and bobs as long as the GDS core remains.
I have looked at the these social GDS’s and wouldn’t give them house room. They are not a GDS in any sense of the word!.
There is a bit of a tendency to favour odd bits here and there. Sometimes on has to gve Amadeus a sharp slap to get it to do what you want it to do and I very often “cheat” that is, check an odd routing on expedia if not happy with the answers I am getting – so it is pointless really, for a GDS to try and pull the wool over an agents eye. AND there’s the nub – it is all very well having any sort of system or feed or app or other toy – at the end of the day, only an agent worth his or her salt can tell if the information being given is correct or garbage. Like the mechanic, the book may say “turn nut A” – but he may know that that is just not right. A lot of saving money in travel is instinctively knowing how to go about something, what “feels” right, what is clearly rubbish (and believe me, there are some very well known travel websites that produce some stunning rubbish). Not to mention the – Ahem! – work arounds. There are other “essential tools” for an agent (or should I say, “I” find essential) one is Google Earth and the other, the website wegolo – which is an excellent aggregator of all the odds and sods low cost airlines.
As I say, I am no techy type. Oh! And my screen is not green. We do have blue ones, now, you know. (or grey…)
- Afterthought – more addressing Valyn Perini – as more and more of the hard nosed travel gets on the web, so too the value of good travel agents increases. A lot of what is available on the web is highly – well, not “misleading” but certainly “in need of interpretation” . The savings I made for one client over a year were very , very substantial. I do find this off the cuff bit about “still do have a value” a little disconcerting. Good agents are more vital now, than ever before.
That said, (though it may sound a bit Orwellian) good agents are getting hard to find – many of the younger breed were simply not brought up using old fashioned methods, so lack (with all due respect) the ability to use their instinct to know when something does not look or feel right – and that can mean buying a ticket for £1,800 rather than £498 (one example of how much I saved on one ticket alone simply by NOT following the airline website’s suggestion.)
Coined the phrase “social GDS” I don’t think its been coined at all. It may be something some marketing type thought up under some misguided intention to try and fool – who – agents? Nope, perhaps the public, I don’t know; but it hasn’t worked and it really is very, very silly. Not clever at all.
The internet will answer your question. But, are you asking the right question??
If you don’t mind me asking why do you even want to define a GDS ?
Are these start ups identifying themselves with a business model or the value (as in cash generative capabilities) of these companies ?
I agree 100% with Murray…top man clearly finger on the pulse!
GDS though spend $Ms in development to support 100Ks traditional agents and Online Travel agents M’s of web consumers. They have to comply with EU and US regulations to provide neutral air displays (hotels and cars don’t fall under this) and a now being required to provision correct lowest priced bookable services to support shopping entries. Rail and air are now being shown together in Europe on GDS primary displays. Try LON/PAR (City pairs) in Galileo instead of LHR/CDG (Airport pairs)and you will see all 9F Eurostar ticketable services including bookable seats – couch 9 seat 34B etc.
The platform is in place if suppliers choose to start interlining with traditional competitng services for example out air back rail on one fare and one E ticket..at the moment these are mainly separate tickets so lossing the traveller option of cheaper round trip fares. These capabilities are avalible via GDSs because they invest in their core business – hi volume transaction realtime bookable content aggregation. Guaranteed bookings for 700+ airlines, 84K hotels, 30+ car companies, LCC content….all in one place – what could be better. As Murray says if you know your stuff – you’ll know competing with the GDSs is not as easy as saying you’re a GDS.
I just tried to grab the RSS Feed for this blog but it is not displaying in Google Chrome. Does anyone have any suggestions?