Executive Profile: Justin Lie, CashShield’s CEO on machine-automated fraud management

Real-time machine learning allows the system to identify fraud before it happens and also automates the entire fraud management process to completely eliminate the need for manual reviews, writes Ai’s Ritesh Gupta


Amidst all the talk around fraudsters finding ways to tame the system, one area where progress is made to stop them is machine-automated fraud management.

Airlines are being recommended to consider the prowess of a decisioning algorithm, with the mandate of maximising revenue at an optimal level of fraud risk. This is expected to make the airline’s fraud prevention methods truly agile at maximising revenue while minimising fraud. 

One company that has been making steady progress in this arena is CashShield. The online fraud risk management specialist has been in news for their $5.5 million Series A round.

The machine learning algorithms used in fraud management today can be divided in two types: supervised machine learning and unsupervised machine learning. Supervised machine learning, or traditional machine learning, involves dumping historical data to train the system, while unsupervised machine learning lets the system learn on the fly with real time data collected. 

So how can one rate the efficacy of machine learning in managing fraud today considering 3 metrics - manual reviews, chargebacks and conversion rate?

Today’s fraud management systems use only supervised machine learning, which is a reactive approach to treat fraud, says Justin Lie, CashShield’s CEO.

“It is true that traditional machine learning has improved fraud management to a certain extent – automating some processes, collecting more data to churn and analyse, but it still falls short from real time machine learning. Without real time learning, supervised machine learning is unable to predict and counter unknown fraud attacks, since it is reliant only on the data on historical fraud attacks. Additionally, these systems can only produce probability scores for each transaction, therefore still requiring a great amount of manual reviews. With the three metrics in mind, traditional machine learning systems are scoring only a 3 out of 10, its full potential,” explained Lie.

Lie asserts that unsupervised machine learning, or real-time machine learning, is far more effective as compared to its supervised machine learning counterpart. Its proactive approach to fraud allows the system to identify fraud before it happens and also automate the entire fraud management process to completely eliminate the need for manual reviews while keeping fraud rates low. Conversion rates also improve with the higher accuracy at proactively identifying genuine customers and fraudsters alike without any additional friction or blocks, he added.

“If merchants are simply rooted in relying on supervised machine learning, it is hard for them to move towards real time machine learning,” said Lie. “Many merchants are also still reliant on manual reviews,  which means that even if they were able to improve their machine learning algorithms and systems, they would always still be held back by the end process of manual reviews and human errors.”

Making the shift

So instead of implementing a fraud prevention strategy that requires long gaps in training machines with data sets, travel companies should shift towards real time machine learning (or real time automated) fraud systems to get ahead of the fraudsters.

Commenting on the sort of progress that has been made as far as real-time automated fraud prevention is concerned, Lie said when airlines and travel companies rely too much on traditional machine learning – and the use of historical data, unknown cyber-attacks without any prior information will pass through the system undetected.

Real time machine learning differs from traditional machine learning in its ability to identify the adversary attempting to trick the system, learning on the fly with each incoming transaction of the fraud attack to match the patterns and block the transactions immediately.

He pointed out that most solutions in the market with machine learning are developed by data scientists who are familiar with traditional machine learning, or merely adopting machine learning algorithms to automate traditional fraud screening methods while still maintaining a heavy reliance on human labour for manual reviews on derived probability scores.

“A probability based approach is basically just predictions that are insufficient to help merchants make decisions. On top of that, cyber criminals are constantly trying to trick the system by feeding manipulated data, making probability measures fall short in its accuracy. This reactive fraud screening methods puts travel companies far away from real time automated fraud prevention, which eliminates the need for manual reviews. Fraud prevention should look to combine multiple disciplines of passive biometrics, financial algorithms, behavioural analysis and predictive analytics to be closer in achieving real-time automation,” he said.

Lie spoke about relevant issues that can impact the fraud management as well as the conversion rate:

·          Rule-based systems: Rules-based systems are in general reactive and probabilistic solutions, which is why they are unable to prevent fraud before it happens. Probabilistic frameworks only seek to train the system on historical data, and do not possess the expertise to move beyond probability scoring for fully automated decisions, thus crippling the system on manual reviews. Because of the need for manual reviews, rules-based systems also start to show cracks at high volumes, and reduces the company’s ability scale on demand. Merchants should be realistic about rules-based systems’ flaws and limitations - mainly on their hindrances to scalability and restrictions to instant delivery.

·          Being wary of blacklists: Lie says blacklists rarely work because hackers will never use the same credit card information twice, while whitelists are inaccurate since whitelisted customers can be compromised anytime. Also, historical data (which blacklists are categorised as) lose relevance very quickly in the face of unknown cyber threats, since it is difficult for the machine to predict new fraud attacks without any prior information. Real time machine learning can help against blanket blacklists and whitelists by focusing on the customer’s behaviour instead. It works with real time live data collected on the merchant’s website, where the system trains itself with each incoming transactions to identify fraud patterns instead.

·          Setting right expectations from new developments: Merchants need to gear up for Dynamic 3DS and Dynamic Authentication in an earnest manner. He says the problem with Dynamic 3DS is that it is controlled by card issuers and is therefore still working with the same set of data as before. They are unable to tap on the merchants’ data for more information on fraud and are not as smart and flexible as they tout themselves to be. Therefore, merchants cannot expect Dynamic 3DS to be a be all and end all solution to solving fraud woes. As for Dynamic Authentication, it is very counterproductive, considering the added friction placed on users. On average, only 70% of dynamic passwords delivered are used, while merchants see a 40% reduction in purchase conversion rates after introducing Dynamic Authentication.

·          Measuring false positives: It is a difficult endeavour as there are many assumptions to be made – that all wrongfully rejected customers actually show their unhappiness in some ways to the airline or travel company or that lesser reports of customers’ calls to customer service is necessarily indicative of lowered false positive rates. “To complicate matters further, many fraud detection solutions in the market like to flaunt their false positive rates (or ability to lower it) to airlines and travel companies, based on ‘internal research’. Instead of auto-block thresholds, merchants should shift away from hard rules and move towards utilising behavioural analysis – looking at the permutations of variables and patterns – which is a much more targeted approach to block hackers and yet minimise false positives at the same time,” said Lie.

·         Counting on data: As explained by Lie in a recent interview, there are two general types of data that one can collect – industry data and unique merchant data. Industry data includes information on coordinated fraud attacks, which may be shared across different airlines as all airlines are equally vulnerable to coordinated hackers. Unique merchant data would vary from airline to airline, based on the individual information each airline collects or is able to provide. He recommended that airlines to amplify and triangulate the data, analysing the data through multiple permutations and combinations so as to better understand the fraud patterns left behind by fraudsters. For instance, passive biometrics data including mouse cursor movements, keystrokes, words per minute or activity data including wish lists, purchase history or even seemingly insignificant data points like whether or not the user has chosen to subscribe to the newsletter can all be relevant information collected and used. With the data collected, airlines can churn the data through various permutations and combinations to identify potential fraud patterns

So airlines should direct fraud prevention efforts on behavioural analysis instead, which is compatible with all various payment methods, currencies and devices. A further step in sustaining or even improving conversion rates for airline can be to develop a decisioning algorithm.


Follow Ai on Twitter: @Ai_Connects_Us

Executive Profile: reward-U’s CEO Steven Greenway on loyalty 2.0

First Published on 8th September, 2017

From no member, no 1st party data to 1 million members in 16 months to gearing up for loyalty 2.0, the journey has been quite an eventful one for Hong Kong Express Airways’ reward-U program, writes Ai’s Ritesh Gupta


What is loyalty 2.0? What is the next loyalty paradigm?

This isn’t mere hypothesis any more. There is action that is emerging now.

Consider the example of reward-U, a lifestyle and loyalty program being managed and operated by Hong Kong Express Airways. Launched in April last year, the program has over 1 million members, serving a growing segment of Millennials. The team asserts that there is a robust foundation in place now (based on dollars spent and not miles flown, few restrictions, any seat any time) with a “liquid currency” (rewarding everyday spending). There are other aspects that stand out even today.  

Take the case of reward-U Crew. Typically everyone in a family isn’t an expert or even interested in ascertaining how to make the most of a loyalty program or spending. But via the reward-U Crew feature, members can set up their own “crew” (maximum of 4 members, plus 1 Captain who ends up being the in the charge of all existing and future points and will decide how to use them). This not only encourages people to let their family or friends join or plan a trip together (since they are spending on non-travel items) but more importantly, it adds a “social” element that today’s consumers are so much used to.

What’s next?

In order to be a true disruptor in the loyalty space, reward-U’s Hong Kong-based CEO Steven Greenway is focusing on 5 areas:

·          Mobile app – an integral for any sort of activity, could be opening an account, customer service etc. The popularity of mobile-based loyalty programs is on the rise. According to study on loyalty in Asia  by Nielsen last year, 69% of loyalty program members in Asia Pacific are more likely to participate in a loyalty program if a mobile app is available, compared to 56% globally.  

·          Gamification – planning to bring lessons from the gaming industry to loyalty. The objective is to “gamify” events, activities etc. in order to stimulate the member activity within the program.  The plan is to roll this out in the next 6-12 months. But gamification isn’t really new, yes today consumers’ have different ways of consuming content, it could be “emojis” for instance on messaging platforms. “Gamification is basically about incentives, everyone loves something “free”.  So a user can ask someone to join the program and gains points for it. It could be for stimulating a certain activity, says surveys featuring members. Yes, it isn’t new but it helps in triggering repeat activity, getting them more involved. We find many people are open to incentivizing and enjoy that type of behavior,” mentioned Greenway.

·          Data-driven  

·          Virtual Currency - high liquidity, minimal restrictions, increase earn / burn etc.

·          The concept of “Tribes” - based on overall activity, for instance, travel, retail, food etc. and a member can be in multiple tribes at any time base on a minimal level of activity. This is scheduled for Q1, 2018.

Tribes – a result of data analysis

“We believe that legacy tiers (as part of FFPs) tend to be one-dimensional. They are too linear base on “status” points etc. typically related only to flying. Tribes is about recognizing your everyday spending and everyday activity patterns. So if a user prefers to go out ,they could be a part of wining and dining tribe. Accordingly, there would be relevant content, plus incentives and promotions to step up the participation as per the user’s activity and the tribe they are part of. That way you have a home, there would be affinity with certain things. We are stimulating that affinity, that activity and one can be a member of multiple tribes – so sports, frequent travel etc. So it’s not one dimensional. This way we are going to leverage the program in a far more effective way,” explained Greenway, who added that the team has been analyzing data, encompassing one million-member base, and as a result has come up with the concept of tribes.

Going forward, it’s not just about segmentation but also 1-to-1 personalisation.

“Starting a loyalty program is always difficult considering that there is no data. But we are 16 months old now, we have been accumulating data, spending patterns etc. We are getting a picture of what people are doing. The association with a low-cost airlines doesn’t mean that our members are low spenders. They are high spenders. They might save money on airfares, but they can spend (relatively lot more) money on accommodation and so forth. Their retailing activities are not low-cost either, they are high-spend. They are bargain hungry but not averse to spending, too.   

“(The tribes initiative) is a direct result of spending pattern that the team has been observing in terms of consumption and activity,” he said. “And it is really about promoting that further.” So this means partners, too, get precise and actionable information about the behavior or preferences of the member base. “How reward-U and its partners will interact will depend on the dynamics within the Tribes themselves. We will overlay this approach with gamification to further incentivise engagement within a Tribe,” shared Greenway.

Being clear with data

So in this case there is 1st party data and 2nd party data about members that is available (according to Greenway, there are 25 partners and the pipeline of partners from airlines, dining, retail, transport, telecoms, travel & leisure exceeds 200; aiming to secure 50+ by the end of year). Organisations tend to struggle with such sources. So what’s Greenway’s recommendation as far as managing data and coming up with actionable information is concerned?

“If you are just looking at data, then you aren’t really doing anything. (In our case), we take specific learnings and then we act on it, and then observe (what happens). The problem is everyone sees data and don’t really take action. We try to experiment from what we are seeing, try to stimulate activity or some type of behavior. Rather than accumulating data in huge amounts, we are taking snapshots and acting on it.  

Follow Ai on Twitter: @Ai_Connects_Us

Executive Interview – Farelogix’s CEO Jim Davidson on controlling the offer

What sort of systems, engines, data etc.are needed to make the most of the product that is owned by airlines? How their ongoing technology contracts limit their own capabilities, and what can be done to improve upon the same? Ai’s Ritesh Gupta spoke to Davidson about the same.


First Published on 26th June, 2017

What can step up the average order value every time a passenger shows the intent of buying an air ticket? How to deliver a sublime passenger experience – one that depicts readiness on the behalf of the airline to answer any question that arises at any stage of the journey via any touchpoint?

In order to optimize monetization as well as deliver a personalized experience, airlines need to move on. Their operational capabilities, the efficacy of their technology and distribution partners etc., everything is under scrutiny. The entire fraternity – be it for airlines, travel technology specialists, the industry body IATA, distribution partners etc. – is in the process of sorting out complex business processes that continue to impede progress.

So how are airlines responding to the need of being in control of their own offering? How are issues related to technology and extraction of data from the current airline-run IT infrastructure being addressed?

Ai’s Ritesh Gupta spoke to Jim Davidson, CEO of Farelogix. Excerpts:

Ai: Airlines are often being singled out for being slow to embrace change for different reasons. They tend to miss out on creating the offer themselves, what they should offer in real-time and not showing the product despite investing in the same, be it for aircraft, meals etc. How are airlines going about countering limitations such as their own IT systems or the archaic process of making an offer?

Jim Davidson: In my opinion, airlines in general are conservative when it comes to change. This is typically rooted in the extreme focus on safety and operations. So when it comes to changing a distribution process where the airline takes over control of creating, controlling, and delivering their offers through an NDC-aligned methodology, it is not that surprising that quite extensive study is required at various levels within an airline. What’s interesting is that many airlines have already completed this initial study. Farelogix observes a certain level of caution regularly during initial discussions with airlines, and this is understandable.

Ai: So when do airlines realize what they are missing out on?

Jim Davidson: The evolution of airline offer control generally begins with a few exploratory meetings about what creating and delivering the offer really means from an airline perspective. Usually this early exploratory stage is driven by maybe two individuals assigned to an “NDC initiative”.

It then progresses from there with more individuals and departments getting involved. This is when you know an airline is serious about taking control of their offer. This is also the time that airlines usually realize that they are either behind when it comes to having the right technology in place, or that they are locked in to someone else’s technology (e.g., PSS and/or GDS) that will not enable them to create, control, and deliver their offers. At this stage, the more motivated airlines look to supplement either their own technology with one of more offer engines, or they confront their PSS with an option to “add-on” a few pieces of new technology to their PSS. This, by the way, is also the point when many airline NDC initiatives get shut down by the PSS, as the airline has neither the clout or commercial ability to do what it desires.

Ai: What is the next step for full service carriers to getting closer to creating their own offer?

Jim Davidson: That is the question. What will they do?

In my opinion, they have two options:

1) Give in to the PSS and hope that the PSS will eventually deliver some capability they need; or

2) Push for a PSS renegotiation that gives the airline more freedom to add their own managed engines that will interoperate with the PSS. Here again the airline faces the obstacles of clout and locked down long term commercial agreements that prevent them from expanding their own technology presence.

The problem for airlines is that they are leaving tens if not hundreds of millions of dollars in new revenue on the table while they wait for their PSS or GDS to catch up. And the reality here is, the PSS and GDSs will never catch up. They simply can’t for two reasons:

1) They are too big and entrenched in existing systems they have in place, and those systems were never designed to do what we want them to do; and,

2) Other companies that specialize in offer engines and distribution-related technologies, like Farelogix, are continuing to rapidly innovate and enhance their offer engines, making it extremely difficult, if not impossible to ever catch up. Some would argue that with enough resources ($$$), the PSS/GDSs will be able to “leapfrog” into a better position. However, in today’s world of development cycles and new technology advancement, “leapfrogging” is a dying art.

Ai: Can you talk about significance of supporting ATPCO-based fares as well as non-ATPCO fares to be managed directly by the airline the way WestJet is going about it? How this needs to be interoperable with other offer engines for off-PSS merchandising, availability calculation, and schedule building? How NDC is playing its part in letting airlines create the offer? 

Jim Davidson: NDC is essentially defining the model and workflow of the airline creating and delivering the offer.

The very nature of the NDC initiative is to create a robust schema that enables any airline to create and deliver its offers to any distribution channel or third party entity.

Having a standardized schema for this purpose makes this a very scalable, repeatable, and over time economically beneficial process. Like any new process change, especially in an industry heavy on legacy technology and institutions, the initial change of process, workflow, and utilized technology requires significant investment from existing players and new entrants. Here, new entrants have a clear advantage over the legacy players, which is why you still see NDC push-back from the legacy players.

Ai: The approach of airlines for selling their core product and even air ancillaries needs to evolve. Can you cite a couple of examples where revenue is being optimized or average order value is going up because of differentiated content that can be pushed via API-led distribution? 

Jim Davidson: There are numerous examples: United Airlines and Air Canada utilize optimization algorithms for premium seat pricing. Westjet uses dynamic pricing capabilities on its Westjet Direct channel. Other airlines use optimized pricing scenarios for upgrades, lounge passes, priority boarding, and more. Farelogix merchandising (FLX Merchandise) and pricing (FLX Shop & Price) engines are used by these airlines.

Ai: What are major challenges as far as interlining is concerned – selling to servicing to financial settlement? Does this hamper direct creation of offer? How are IATA’s initiatives taking care of such complexities going forward?

There is certainly a new level of complexity when it comes to offering additional content from interline partners. I do not believe that these complexities hamper airlines wanting to directly create their offers. It seems that the logical adoption flow is starting with online selling and servicing of additional content, moving into the alliance with these capabilities, followed by unaffiliated interlining. It will obviously take a bit of time, but IATA is doing a lot of work in defining standardized workflow processes and schema elements to enable NDC interlining capabilities.


Hear from experts at the upcoming Controlling Your Offer Symposium, featuring live demonstrations that take NDC, dynamic pricing, revenue management and merchandising to the next level.

For more click here

Follow Ai on Twitter: @Ai_Connects_Us

Executive Interview: JR Technologies’ Ryan Harris on what is impeding “transformation”

First Published on 24th April, 2017

Airlines need to look at several areas, including the organizational impact, IT, issues associated with access to data and their innate nature to detest change, in order to embrace digital transformation.


Where do airlines stand today when we talk of delivering a seamless, relevant experience throughout the passenger journey?

Ai’s Ritesh Gupta spoke to Orlando, Florida-based Ryan Harris, Director of the Americas, JR Technologies, about the same:

Ai: Can you explain what sort of transformation is required by airlines – IT infrastructure and APIs - to gear up for offer management, order management and customer management in the best possible manner?

Ryan Harris: There is not an easy answer to this question, mainly because the operations and business goals of each airline is different than any other airline.  But, it starts with internal education.  The stakeholders within the entire corporate structure need to know what is coming and how it will affect not only their own business line, but also all the other lines in the company. 

Next important step is to set your organizational goals and diligently plan how you are going to get there.  Some of the biggest difficulties I see for most companies won’t be the technical changes, but the organizational changes that come with that through the changes in required business functions.  For example, with a complete transition to NDC distribution and One Order, there is no longer a need to file fares.  What do you do with your fares department, which is sometimes a rather large asset of human and information assets?  Do you just cut the labor and salary cost?  Do you transition them to other departments?  If you transition them, how do you plan the required training while they are still providing a business-critical function?  These are the types of issues that I see as a much bigger challenge to address, not just the technical side.

But, of course, you can’t just forget the technical side.  This is an industry transition, not just a company, so while you may be fully ready to jump in and get into the Offer-Order management pool, your interline partners will likely still require the legacy connectivity to be maintained and you will still need the legacy systems to distribute through traditional GDS channels until a critical mass is built in the NDC Aggregator channel to replace it.

Ai: Flydubai has worked out end-to-end PSS API in public domain. How do you assess moves like these when we talk of collaborative approach towards serving passengers in the best possible manner?

Ryan Harris: Public APIs and the entire open-source community, I personally think, have been some of the greatest innovators in the digital revolution since the beginning. 

Making things public and open generally leads to exposure to people outside the box.  They can think of things that some of us in the industry just can’t conceive.  Keeping things closed forces tunnel vision and can stifle creativity. 

I think the industry is starting to come around a little bit on releasing public API’s, at least high-level.  For example, the NDC, and eventually One Order, XML schemas are currently available for public download through IATA.  This means that anyone can get them and create any sort of interface that they want, a fact that IATA itself has leveraged through its several Hackathons.  I’ve been to a few of them now and it’s interesting to see how people from outside the industry want to interact with the industry.  Most are good, some are questionable, but there are a few brilliant ideas that come from some of these participants.  Events like that are only possible because of the open-source approach that IATA is taking on the industry’s behalf.

Ai: How can the power of APIs enable airlines to reach out to audience at the right time via the interface they chose to interact with? For instance, for instance, if I am chatting with a friend on WeChat or Facebook Messenger and planning my next flight, how can a link for search and booking be worked out by various partners via APIs? 

Ryan Harris:That is very much up to the limits of the individual airline, which may be decided by either the willingness or ability of their system providers.  Some of the airlines with the greatest ability in this area are the ones that run their own systems completely and can do whatever they want to do with it.  The problem is, there are very few airlines in the world that either want to take that level of involvement or can afford to do it.

But, this also gets us back to the question of the open API.  The airline industry is, by its competitive nature, very adverse to openness.  We are also, by our safety-conscious nature, very resistant to change.  An airline or a provider can fully publish interactive APIs to the public today, and it would likely be picked up and built upon in creative fashion.  But, I remember a few years ago, there was a little bubble for apps that could consolidate all your loyalty points and show you in one screen what your balances were and I found it quite useful.  But then one airline blocked access, then another, and another, until eventually there was no source for these apps to access.  The reasons varied from security to lack of brand exposure, which after all, is the actual goal of a loyalty program, but the result was that the airlines removed that data source from the public domain and the innovation that was built on that data was quickly destroyed.

So, can something like this be done today?  Probably.  Will it be easier tomorrow? Likely.  But it’s going to be up to someone in the industry to open the door and let the people in to do it.

Ai: So, how can airlines serve passengers in a seamlessness and relevant manner throughout their journey?  Can the blend of data, cloud, IT, content and emerging consumer technologies propel ancillary revenue?  

Ryan Harris: Knowledge is power.  There are entire industries based on nothing more than the data surrounding an individual person’s movements and economic activity.  There are likely a lot of things that would be completely possible from marketing, promotions and sales prospective through the data available in today’s mobile devices and the Internet of Things.  There are only three issues that I see causing limitations, and technology is not one of them. 

The first is consensual, as in most countries, companies are required by law for you, as the consumer, to opt-in to allowing the collection of that data by a company.  The second is legal, as again, most companies are limited by the law as to what they can do with the data you allow them to collect. 

To be honest, the third limiting factor, in my opinion, is what I call the “Creepy Factor”.  I still get a little skeptical when an idea or a product that I was just talking about in a face-to-face conversation shows up in an ad on a website.  Maybe it’s just coincidence, but maybe not, I don’t know.  But, the fact remains that most people are not prepared to fully give in to the artificial intelligence big data machine telling us everything we want when we want it, even though it is pretty much possible today.  If you build in too much of this predictive AI, you run the risk of spooking your customers and seeing exactly the opposite result that you were looking for.

So, yes, it is possible, but with some careful restrictions and customer education.


Gain an insight into intriguing issues at Ai’s 11th edition of Ancillary Merchandising Conference in Spain this week. 

Date: 25 Apr 2017 - 27 Apr 2017;

Location: Mallorca, Spain 

For more info, click here

Follow Ai on Twitter: @Ai_Connects_Us




flyiin’s Stéphane Pingaud on disruption via API aggregation

First Published on 27th March, 2017

flyiin is 100% API-based online marketplace that promises to show travellers offers that are relevant to them, and in doing so provides airlines entire control over the distribution of their product, writes Ai’s Ritesh Gupta


How comfortable is an 18-inch economy seat? What is the meaning of “customisable lighting” in an aircraft? What sorts of ground transportation options are available at a certain airport?

These are just a handful of questions that a passenger might contemplate while booking a seat on an air plane. And considering the “fickleness” of a consumer in the digital world, how can airlines present their product in the best possible manner when one is looking for the same?

There is no point in making investment in a new aircraft if the same can’t be conveyed to the traveller and that too in a manner that adds to the joy of flying.

In this context, digitisation of operations is rightly being strived for, by drifting away from being technology-centric processes that this industry has over the years followed. Collaboration via APIs is one route that entities, including airlines, are increasingly embracing, and this means the distribution status quo is set to be challenged.

Also, if more airlines are pushing their offerings via APIs, what sort of opportunities is there for start-ups?

Berlin, Germany-based flyiin is the process of crafting a new online marketplace that is 100% API-based, playing the role of an aggregator and working out a new sales channel.

The company recently signed a pilot agreement with Lufthansa Group to take part in the beta phase of their offering. The start-up doesn’t rely on GDS technology. Rather the team has developed their own search and booking technology, plus it is capable of aggregating APIs from the airlines. And even if there is lack of standardization in APIs, flyiin would work on the normalisation of APIs, and then facilitate search and booking (via request and exchange of data, could be content or inventory). So essentially flyiin would exchange data as per the version/ interpretation of the airline, and eventually transform the airline’s response message to NDC 16.2 version. Till recently, the platform could normalise APIs from six airlines, based on different versions of NDC (1.1.3, 15.2 and 16.1).

Ai’s Ritesh Gupta spoke to flyiin’s CEO Stéphane Pingaud about the status of the venture. Excerpts from the interview:

Ai: How does flyiin compare with existing online intermediaries such as online travel agencies (OTAs) and meta-search engines (MSEs)?  

Pingaud: There are quite a few differentiating aspects to OTAs and MSEs but the most relevant one for the consumer is that the search - and subsequent booking, payment, ticketing and servicing transactions take place directly with the airlines. When you search for flights (and services) to your desired destinations, your search is sent to all airlines operating the route directly or through their hub. That means the flight information, the fare and any fees for extra service are returned by the airline. Your search result information will always be valid and fares available. Not always the case with OTA or MSEs.

Secondly, as importantly, you will get the visibility of the full costs of your flights prior to booking them. If you search for flights between London and New York, and wish to check 2 bags, select your own seat and be able to change or cancel your flights if needed, flyiin will return the total costs, inclusive of these services. So you can easily compare flight options irrespective of the various bundled fares and ancillary services approach of each airline operating the requested route.

Ai: Why are you calling flyiin a new sales channel?

Pingaud: So flyiin is an online marketplace for air travel, where airlines and travellers interact directly throughout the entire flight planning and purchase (and post-purchase) process. There are no technical intermediaries between the airlines and their customers. Like any marketplace in other verticals and industries, for airlines our sales channel is semi-direct. They keep the entire control over the distribution of their product throughout the whole purchase process.

Ai: What’s your vision for flyiin, and how are you looking at filling the gap in current planning and booking?  

Pingaud: flyiin aspires to be the number one brand and product for online flight planning and purchase. How? By bring the both the comparison power of OTA/ MSEs with the in-depth content of airline channels. But once again, by minimizing the complexity of individual airline offerings and making it easy for travellers to compare these offerings in one screen, including of the services that are important to them and available in each flight option.

Ai: Can you elaborate on revenue generation? Also, how do you intend gain traction, looking at growth hacking?

Pingaud: I can't elaborate on the revenue generation at this stage, as we will use the 16-18 month beta phase to get the data that will help us define a proper business model with the pilot airlines. Although new for airline distribution, our business model is typical of a marketplace. Travellers will search, shop, book, pay and get their tickets directly with the airlines. They will be in the flyiin digital environment (like Amazon) but in the background, airlines through their APIs will be in charge. It goes beyond facilitated bookings and encompasses 'facilitated search'.

About growth, the next 18 months - financed by our seed round - are going to be all about building the best possible marketplace product. Growth hacking will be after we secure series A.

Regarding white label and other user channels like messaging channels etc., yes we will consider all of those, to make sure that travellers can use flyiin wherever they are, but once again right now of our focus is on building the core product and supporting API aggregation platform and get as many airlines as possible on-board.

Ai: How has flyiin come up with a technologically advanced, NDC-based distribution model?  

Pingaud: We knew the only way we could really deliver a much simpler, more transparent and user-centric experience to travellers was to disregard existing search and booking technology from the GDS (and other fare search system providers) and instead leverage the APIs from the airlines (many of which are based on the NDC standard). As a consequence, we built our API aggregation platform which connects to each airline API and 'normalise' these APIs into the latest version of the standard, since all these API are based on different versions of NDC (or are not NDC-based) and/or interpret the standard differently.

Ai: Can you explain how are you looking at commercializing all offerings of airlines, be it for core offering like air ticket, or air ancillaries and even bundled fares and other ancillary services?  

Pingaud: Our approach to flight comparison is to ensure that travellers can compare all flight options returned by all airlines being queried, inclusive of the services that are important to them (number of bags for check-in, select my own seat, flexible fares etc.). We do by showing the fare from the right fare family i.e. that includes the requested services.

The plan is to also associate to each flight option a 'flight details page' that will enable airlines to showcase the experience to be expected during that flight, using a combination of media assets and product/services descriptions.

Ai: Can you explain how flyiin would contribute in terms of real-time data exchange that can help airlines to push the right offer at the right time as per the context or intent of the traveller?

Pingaud: For every search made by travellers through flyiin, airlines receive an XML search request messages that will include not only their desired O&D and dates and number in parties, but also the services that they wish included with their flights. If the traveller is logged in, they will also know who is searching for flights and adjust their offer accordingly. Secondly, we’re using a technology for our front-end development that can potentially provide data about what travellers are interested in (services, destinations etc.) which we would potentially shared in real time and anonymously, but it is too early to talk about this.

Gain an insight into intriguing issues at Ai’s 11th edition of Ancillary Merchandising Conference in Spain this year. 


Date: 25 Apr 2017 - 27 Apr 2017;

Location: Mallorca, Spain 

For more info, click here

Follow Ai on Twitter: @Ai_Connects_Us

Dean Dacko on fostering a personalised customer relationship

First published on 7th March, 2017

What are the major obstacles to fostering a personalised customer relationship?   

Experienced airline executive Dean Dacko asserts the challenge is three-fold: the lack of a full 360 view of the customer behaviour including customer touch points outside of marketing + competitive search and transaction activity; the inability to convert customer data into actionable customer insights to deliver relevant customer value in real-time; the fear of making the major investments required (technology, systems, processes, talent) without a guarantee of return on investment success.

Current challenge

Data is housed in specific silos across the organization. It doesn’t allow for a single view of the customer. So how to interact with customer, how to communicate with them across various points of their journey?  Customers are ahead, and airlines aren’t able to deliver personalised relationships that entities in a lot of other customer verticals are already delivering. It needs to be noted that there could be brands that belong to organizations that aren’t more than 15 years old. They probably started with a platform and business proposition that doesn’t entail involvement of a legacy system. But recognize that their whole ability to succeed and grow lies in personal relationships. So the customer is expecting that the brands they appreciate the ones they want to build a relationship with and the ones they want to transact with are the ones that truly understand them and deliver a value proposition that’s unique to them. In the context of understanding what a personalised relationship looks like - it is about delivering personalised customer value proposition unique to customers. This makes customers feel “You get, you understand me, you anticipate my question before I pose it and you answer/ present me with a relevant offering”. So when talk of that moment of truth, customers always find that brand to be in the scheme of things as per their booking funnel. “Understanding is important rather than offering free stuff,” said Dacko. “So it’s the notion of using personalisation in growing that capability to create a personalised relationship that makes them believe that you are truly investing in them. This reflects in the trust that is developed. So the notion of where the future is going in terms of understanding the importance of personalisation is really something that is understood how the customer is evolving not just in the airline space, but almost every other vertical worldwide.  


Gain an insight into intriguing issues at Ai’s 11th edition of Ancillary Merchandising Conference in Spain this year. 

Date: 25 Apr 2017 - 27 Apr 2017; Location: Mallorca, Spain 

For more info, click here

Follow Ai on Twitter: @Ai_Connects_Us

Ai Editorial: Behavioral biometrics, detecting a bot…fraud prevention isn’t same anymore

First Published on 17th January, 2017

Ai Editorial: Travel companies aren’t new to tracking suspicious online behavior, but time has come to refine the ploy to make it more sophisticated as well as category-specific, writes Ai’s Ritesh Gupta  


Certain level of sophistication, irrespective of the fact that whether the fraud prevention method is being used for the retail sector or an airline, is definitely needed. So, for instance, machine learning is able to detect certain patterns that can be termed fraudulent. There could be a couple of red flags in one session, say password change and change of shipping address, that can differentiate authentic shopping flow from an illegitimate one.  The tracking of one’s navigational footprint can give ample indication whether a fraudulent transaction could happen.

Specialists point out that fraudsters need to be “out-smarted”, as tools and technology need to spot something that fraudsters wouldn’t think about!

Here are few areas that airlines can focus on to improve further, by acting on moves/ patterns that can be spotted on their digital assets. These initiatives not only combat fraud independently but also combine to make the whole effort even more fruitful: 

·          Do act on data that useful: Existing fraud solutions are designed to cater to mass markets where most airlines will only need to collect data based on a template that analyses very limited fields. This is not only insufficient, but also limits a merchant’s ability to create an optimal data strategy and reporting for their performance/ ROI. Unfortunately, not much useful data is returned to the merchant by default.

As each airline’s ecommerce website is unique, the data strategy deployed must be different and customised.

It is important to work with airlines and help them utilise all the data that is available on their website. Some custom data fields that may be collected include: flight details, loyalty miles claims (to detect abnormalities), or even a small, seemingly insignificant data field of whether the newsletter subscription box was checked or not.

·          Tracking behavior for authentication:  Behavioral analysis is one area that is becoming increasingly sophisticated. Swipes, taps, cursor movements etc. are being analyzed for navigation flow, time spent etc. to understand the behavior. Specialists are tracking mouse movements and clicks in context and meaning while becoming increasingly more accurate over time. User data is important to understand the user behaviour, for instance the words per minute (WPM) typed, how the cursor moves around the website, existing patterns of the card user, rather than simply focusing on the card blacklist or whitelist.

Visa, in one of its recent blog posts, emphasised that organizations today require a holistic approach—“one that begins by reducing the threat of fraud when the customer first establishes an account and continues all the way through the moment an online transaction is approved”. The company adds that a multi-layered fraud management approach is must. The goal of airlines should be – monitor each visitor, creating a unique device profile that accrues the device’s history over the Internet. This device information is associated with behavioral pattern exhibited by users. Further this is analyzed and compiled over a period of time, and then the real-time rule-based decision-making based on transactional data, in conjunction with device and behavioral data, for acceptance or rejection of a transaction.

It is also being suggested that behavioral biometrics, which spots patterns in human activities, needs to be looked upon for continuous authentication, and looked beyond the two-factor authentication (2FA) method. 2FA is a ploy used to make it tough for hackers to gain access to a user’s devices/ online accounts. So by just having a password one cannot clear the authentication check. Plan is to protect data from hackers who have stolen a password database or used phishing campaigns to gain users’ passwords. Speech pattern, ID card etc. is the second layer here. But it is being recommended that organizations now need to go for stringent processes that persistently evaluate and check the authenticity of users that are intricate to reproduce. The industry is making progress to precisely validate user identities via their inherent and subtle interactions online – behavior that cannot be imitated by a 3rd party.

So with more and more data analysed, it is harder for hackers to hide their tracks fully to pass off as genuine. By identifying user behaviour (between a genuine customer and a fraudster), fraud rates and chargeback rates will fall when fraudsters are effectively blocked by the fraud system.

·          Protecting customers: As we highlighted in an article last week, every piece of customer data and information is under scrutiny. One can put a price tag on stolen account info – Uber, Facebook etc. and air miles. Yahoo, LinkedIn etc. have struggled of late in this arena. When the user account on one airline’s system is breached, hackers will use the exact credentials to take over the same user’s account on the other airlines’ systems as users seldom differentiate their login credentials. Travel e-commerce players should also move fast to be ahead of the curve and protect themselves against account takeovers. Here also there is a need to identify anomalies in real-time, and specialists are assessing behavioral data points to determine if it is the genuine account holder or an imposter.

Hackers/ fraudsters require automation, and rely on botnets to input user credentials. So how to detect a bot? Here Captcha, putting a limit on the volume of traffic that can visit a site during a given timeframe, fingerprinting etc. can help.

The variety and rising speed of fraud phenomena is forcing airlines to move swiftly. Be it for data or new technology, it time’s to look beyond the so-called mass-market or traditional solutions. 


Are you bold enough to survive in the brave new world?  Assess your preparedness at 11th Airline & Travel Payments Summit (ATPS).

Date: 03 May 2017 - 05 May 2017   

Location: Berlin, Germany 

For information, click here


Follow Ai on Twitter: @Ai_Connects_Us

Executive Interview: Transpay’s Mike Carlo on cross-border payments

First Published on 5th January, 2017

Every market has its own payment mix, mandates, checkout and risk management considerations. When you think about the scale of global travel, monitoring these multinational regulations is a rigorous process.


Payouts are a complex process that entail one organization making mass payments to different recipients and bank accounts. With payouts, merchants have the opportunity to manage for cost reduction and efficiencies. A merchant can manage the process to be in their best interest versus with a consumer conversion that requires more flexibility.

As being witnessed with payment acceptance, travel companies are embracing an array of cross-border payment options for a local payout experience.

This is a welcome change as cross-border payments have progressed rather slowly over the years, mainly owing to the fact that one has to adhere to stipulated regulations. In the past, one had to deal with issues related to fees, long payout timelines and archaic banking networks. While initiating cross-border bank transfers, travel organizations encountered slow transactions and opaque funds flow. This was because funds had to go through multiple financial institutions to get to the ultimate end recipient. Each stop along this correspondent bank network also comes at high cost, as each financial institution charges a fee for handling the transaction. But now travel brands are increasingly trying to process payouts locally, bringing down the number of financial institutions involved, and ultimately reducing the cost of sending mass payouts.

“Instead of relying solely on standard global payment services like wire transfers, expect 2017 to boast new global payout methods allowing suppliers to simplify cross-currency transfers using local currency funding,” says Mike Carlo, Global Head of Travel Payments, Transpay.

Among other major trends for 2017, Carlo expects cost-management to gain prominence among travel companies planning to rationalize international funds transfers. He further added, “(Also) as the world becomes more interconnected through new technologies, manual invoicing practices will gradually decline across the industry. From the elimination of fax for hotel bookings, to refund payouts for distressed airline passengers, the travel industry is eagerly moving towards more scalable payout solutions to automate settlements.”

Ai’s Ritesh Gupta spoke to Carlo about cross-border transactions. Excerpts:

Ai: How do country-specific licensing requirements or developments such as Brexit and volatility in the Chinese market impact cross-border transactions? How such developments impact cross-border exchange for merchants, PSPs and consumers?

Mike Carlo: While it’s still too early for to know the full impact of Brexit, current market and currency volatilities, as well as potential government imposed sanctions, illustrate the need for merchants to pay close attention to the regulations associated with cross-border payouts.

Each country and financial institution has its own set of regulations, including anti-money laundering (AML) policies, which are frequently updated. When you think about the scale of global travel, monitoring these multinational regulations is a rigorous process and often requires outsourcing to companies with a dedicated staff, tasked with ensuring compliance.

Ai: Every market has its own payment mix, mandates, checkout and risk management considerations. What peculiar developments would you like to share as far as cross-border payments is concerned? Can you talk about this in the context of a market like China?

Mike Carlo: While not as expansive as payment acceptance, there are several complexities to supplier payouts, specifically in the area of cross-border transactions. For example, there are limitations to sending payments to individuals in markets like China, which has tight restrictions on the available pay-in options for the country; preferring instead to rely on its own currency.

From a B2B perspective, there are few truly global options, and merchants are traditionally left with costly electronic funds transfers like SWIFT and bank wires that take a long time to settle. Unlike payment acceptance, payouts are not about conversion but instead focus on cost management, so anyway to localize the payment experience and reduce cost, is critical to streamlining the process.

Ai: Funds sent internationally are subject to a number of regulatory and compliance standards. Can you explain what these standards are all about? In case countries lack requisite standards needed for movement of money, how can it impact business?

Mike Carlo: In the travel sector the main compliance standards focus around “knowing your customer” (KYC) and AML policies. Considering the number of global transactions occurring in travel, the industry wants to ensure it’s not aiding money laundering or supporting terrorism. For example, AML risks can arise when illicit fund transfers are masked as a hotel booking or a ground package purchases to small and unknown suppliers. To avoid this risk, financial institutions and companies like Transpay will check recipients against watch lists like OFAC OSC.

Ai: Talking of an area like airlines issuing refunds on cancelled flights or OTAs need a payout option for making commission payments, how can travel companies handle such routine transactions in an optimal manner?

Mike Carlo: The best practice would be to partner with an organization that manages, long-tail and to challenging markets. Merchants with significant presence in a few markets can oftentimes handle payouts in those markets themselves (opening a local bank account etc). However, the infrastructure and management required to process payouts makes it virtually impossible to scale this for every country.

Ai: Overall, how is the travel industry looking at real-time payouts in the B2B arena?

Mike Carlo: They are not but we’re looking to change that in 2017. If you look at the competitive nature of online selling, travel payments has focused on getting the largest amount of transactions in the door, and have spent the last 10 years fine-tuning the acceptance process. However, for every dollar in, there is a dollar out and very few companies have investigated to cost of sending payouts.

Ai: Can you talk about challenges that arise when applying e-wallets to global mass payouts?

Mike Carlo: The biggest challenge is eWallet acceptance and costs associated with it. Wallets are largely a consumer-based solution, so they are a viable solution for B2C transactions but not when applied to B2B payouts.

Ai: Can you explain how fraud risk on cross-border sales, which tends to be higher than domestic e-commerce transactions, has shaped up?

Mike Carlo: Fraud is much more of an acceptance challenge. Payouts are merchant pushed, so fraud is not part of the equation. The bigger issues in payouts center on AML and KYC risk.


Ai is set to conduct the 11th Airline & Travel Payments Summit (ATPS) this year.

Date: 3 May - 5 May 2017   

Location: Berlin, Germany

For more info, click here


Follow Ai on Twitter: @Ai_Connects_Us

Executive Interview: Triometric’s Jonathan Boffey on leveraging XML shopping data

First Published on 4th November, 2016


An omni-channel approach, with consistency across all channels, is the most efficient, most profitable distribution strategy for airlines. But airlines have to make diligent moves to attain such level of proficiency.

Data-driven merchandising and API-led distribution are laying strong foundation for a carrier’s retailing strategy, optimizing performance of both direct and indirect channels.

Not only are airlines re-looking at their whole IT infrastructure, but they are also beginning to recognize the significance of a business intelligence layer that can pave the way to regaining control of the offer being made via the indirect channel. Essentially this means airlines need to delve into search and booking data contained in XML message streams. By doing so, one can understand customer groupings in terms of customer type, degree of engagement, revenue contribution and the ancillary upselling-purchasing opportunities, says Jonathan Boffey, SVP for Business Development, Triometric.

Ai’s Ritesh Gupta spoke to Boffey about XML analytics. Boffey, who was in Toronto last week for Ai’s MegaEvent16, asserted that airlines can capitalise on shopping data from intermediaries, and reap benefits in several ways, be it for their IT operations, availability of inventory, pricing strategy or product relevance. Excerpts from the interview:

Analytics is paving way for effective retailing. Where do airlines need to improve in order to be in control?

Jonathan Boffey: Airlines, like any e-commerce brand, are earnestly trying to garner differentiated competitive advantage in today’s era of personalised retailing.

Customer insight is of paramount importance as it offers the ability to engage with context. A prospective traveller might search directly on airline.com, complete a transaction within Google’s domain, an OTA platform etc. The challenge is getting the right offer to the traveller at the right time and through the channel they prefer. Consistency and personalisation is key.

So it is vital for airlines to know who is looking for what, whether they are buying or not and if buying what they are buying. But if this is only restricted to a direct channel, then it isn’t an ideal scenario in the omni-channel shopping environment. So as much as airlines today can minutely scrutinize every piece of activity on their digital assets, they need to strengthen their analytics for indirect channels, too. Airlines have control and visibility when we talk of direct sales channels. But this control and visibility also needs to extend to indirect sales.

And with NDC, if a carrier isn’t analysing then it’s a missed opportunity, right?

Jonathan Boffey: Yes.  

Without NDC, airlines have no visibility into search traffic and therefore into how they are performing nor can they sell the same range of value ancillaries at the same time as the seat sale. NDC is all about bringing control back to airlines in terms of offers they are able to make via this channel. Therefore, this channel’s performance should be able to be analysed and scrutinised with an analytical lens in the same way as the direct channel.  

So what needs to be done to sharpen overall merchandising strategy?

Jonathan Boffey: Travel brands need to engage customers using decisions formulated from customer contact and transaction data.

Increasingly airlines are counting on XML to deliver content rich offers to OTAs, aggregators etc. What needs to be added to this new way of merchandising is a business intelligence layer. Essentially this means airlines need to delve into search and booking data featuring in XML message streams exchanged between airline systems and travel agencies. The key is to act on this in near real-time.

By sorting out XML search and reply data streams and measuring them against key performance indicators, valuable insight can be derived to help shape merchandising /offer decisions.   

One of the biggest assets that airlines’ today possess is the offer they make. Analysing data flow can shape up the offers they intend to make, and this would in turn differentiate their proposition in the indirect environment.

We are equipped to capture the data, offer storage and processing capabilities, plus a web-based interface and reports and charts. All this can be scaled up, too. As a BI and performance monitoring platform, the Triometric Analyzer can be deployed as part of an organization’s data centre capabilities or via a preferred merchandising and distribution platform provider.



What does shopping data feature?

Jonathan Boffey: The data that is coming in via request from the indirect channel shares what is being demanded. Stored in XML format, it encapsulates the actual product search, the traveller or party, the date of travel and other attributes. And when an airline responds, it entails details of features, price, availability and ultimately booking confirmation. Now the analytics platform assesses each request-response pairing, applies NDC specific rules to pull out business relevant data as KPIs and stores them in a database. This way the gist of the transactions are captured and the search patterns are analysed. What effectively surfaces is trends by source - where requests are coming from tends to influence the nature of the enquiry such as destination, check-in date, itinerary, leisure/family etc. It becomes easy to generate reports showing for example conversion rates by agent and/or by destination by check-in date or indeed any phase in the plan-book-travel journey. Ultimately the data available for analysis and reporting is dependent upon the data content of the original XML – which fortunately is pretty good.

How can airlines build on their resources with such insights?

Jonathan Boffey: The key to unlocking the potential of today is identifying the quick wins across the business to ensure incremental and impactful changes, as well as keeping an open mind about what is possible in the future. Internal alignment is necessary, and organizations need to be nimble when it comes to leveraging insights derived from XML shopping data.

We believe airline business intelligence is all about breaking down those silos and barriers – and in key areas we believe XML analysis can play a critical part. Most have data practices across areas like revenue management, marketing, network planning and inventory management. However, it is often the case that these applications are not fully integrated into other parts of the business and are often poor at sharing data. For example, when it comes to things such as ancillary sales and customer data, this is often not in the hands of the revenue managers in time to help influence their pricing decisions. 

Airlines should be in control of their business on various counts:

-       The best performing routes

-       Best customer groups and what do they look like/ their preferences

-       Most relevant ancillary cross-sell or up-sell by customer group/ route

-       What’s the booking curve?

-       Identify customer groups and/or products fuelling growth, and ones on decline

Other than understanding what’s being requested and evaluating which customers are most valuable, airlines can plan several initiatives such as market specific promotion, and when to start or stop promotions; continuously refine flight schedules etc.

Follow Ai on Twitter: @Ai_Connects_Us

Executive Interview: Michael Strauss on controlling inventory, distribution and sales

First published on 4th October, 2016

How can airlines bring about change – being in control of their IT, offers they make, content they show, distribution etc.? Is the current landscape that features tightly integrated processes too rigid to crack easily? Ai’s Ritesh Gupta spoke to PASS Consulting’s Strauss about the same.


Airlines are keenly looking at emerging options, be it for their IT infrastructure or indirect distribution. Two aspects clearly stand out as of today – the pace of change and the quest of being in control. As witnessed, there are new arrangements that are emerging. For instance, Siemens, together with Lufthansa, started using a bespoke Direct Connect Solution for Lufthansa Group in August. The new arrangement uses Amadeus Altea and Amadeus- owned Cytric OBT, with GDS being bypassed.

Considering that Amadeus and Sabre are deeply entrenched in this space, what can one expect in the future? 

In order to know more, Ai’s Ritesh Gupta interacted with Michael Strauss CEO PASS Consulting and Head of PASS Business Unit Travel. Excerpts:

Ai: Where do airlines, cumulatively as an industry, stand today in their quest of being in control of their inventory, distribution and sales?  

Michael Strauss: Nowhere! With the exception of a few airlines, most of them are at the mercy of their PSS or CRS provider. With a couple of mergers (American Airlines or AA and US Airways or Continental and United Airlines) we have seen what an undertaking it is to change the inventory provider (or even merge them) – a multi-year project costing a fortune. So inventory management systems are definitely in the driver seat.

Airlines are smart if they at least put sales and distribution in other baskets – which most of them do with their own website or direct connects with large sales organizations (e.g. Priceline) or corporations (Siemens). But considering dependencies, we were able to witness how hard it is to change the landscape: AA in my eyes totally failed with their initiative a few years ago and only a few island solutions remain, and Lufthansa’s success remains to be seen - and it was only possible for them to get this far due to their domestic power. In sales sector business is underestimated – the TMCs maybe do not issue the majority of tickets, but bring in the most revenue. And this is a tightly integrated sector where things collapse if you try to change a screw. I believe TMCs have put too many eggs in GDS baskets and with the constellation that most inventory management systems are owned by GDSs this makes for a dangerous combination: GDSs control the whole value creation chain (image below) not just with their own sales organizations, but also with their influence on TMCs.

So all this friction makes sense to break some rules, but to use a hammer like Lufthansa did might not work for everybody.


Ai: Talking of direct connect strategy, how successful airlines are in their endeavour of being in control and also offering corporate travellers a service tailored to their individual needs?

Michael Strauss: It is lacking, there are so many tightly integrated processes that I’m not surprised. I still believe the strategy is not to change it all, but rather to make organizations think that they need to reinvent themselves. Technology may certainly help with providing individual needs, but even for us at PASS - we have all GDS integrations and a Lufthansa integration - it wouldn’t make sense to do the same for 480 airlines. So while technology can enable certain disruption to make people think it cannot change the whole landscape and all these players are in there for a reason – it is just that too many players are trying to prevent the status quo and block innovation. But as long as the majority of all the inventory is controlled by Amadeus Altea or SabreSonic there will only be change to a certain degree. It would have been interesting if Google used ITA’s res initiative to provide an alternative res system – but then you would put yourself into Google’s hands which isn’t much better. Different story when it was still ITA software.

Ai: It is pointed out that ancillary or bundled products are not readily available when shopping through most travel agencies or corporate booking tools – creating an unnecessary discrepancy and lost opportunity for the airline. What do you make of the situation?

Michael Strauss: Personally I’m annoyed by this topic – and also considering it a ‘lost opportunity’ which I believe means lost revenue – there is not lost revenue. People are ready to pay an amount for their trip and if it is too much and there is no alternative they don’t go. What changed? Checked bags cost extra – in the past it was included. It was considered unbundling, but if it was unbundling, the ticket price should have dropped and since so many travellers started to pack lighter and don’t check bags the airlines should have actually lost revenue – but they didn’t. So checked bags is nothing else than a hidden price increase. And so is seat assignment, etc.

But this is just beside the fact. What I’m really annoyed at is that we are keeping ourselves so busy with technology and discussion on which channel gets what and so on, and no one looks at the user experience at all. The shopping and booking process is more complicated than it ever was. No matter where I book I always get a price which isn’t tailored for me - no system gets smarter with my preferences and search habits. I cannot really shop with my preferences. Don’t even bother showing me results on a middle seat, a connection at an airport which I don’t like, an angled business class seat, fares with outrageous rebooking fees, codeshare flights which don’t even allow me to book a seat, fares which do not satisfy my mileage expectation, etc. -  all we can search today is basic price and/or schedule - and this is just made more complicated for the user as now he has to pay extra for a seat, for luggage, etc. on one airline which he doesn’t due to his status, policy or other things on another one – this is not even possible within the alliances. There is so much more we could do if we all just stopped the fighting and start innovating. As per my chatbot blog: I want a fair offer at a fair price for what I need including my preferences and this shouldn’t come in 20 steps, it should be right at your fingertips. (In his post on chatbot, Strauss refers to utility of a chatbot. One could possibly request for a change in a travel plan, and how right from knowing the PNR to details of the whole itinerary, chatbots could possibly help with not only mundane work but also proactive decision-making say intelligence related to destination you are going to).

Ai: Airlines need to work on a standardized API that would be consumed by all channels – web site, kiosk, GDSs, mobile, etc. Even as airlines are trying to be in control of distribution, even IATA points out that API XML connectivity of certain airlines is being done in a proprietary way. Where do airlines need to improve to ensure all stakeholders in the travel distribution chain benefit as well?

Michael Strauss: Indeed the problem is one standard. We have gone through this in the year 2000 when we were part of the Open Travel Alliance and were in the process of developing a Multi-GDS interface. The problem with standardization organizations is that there are numerous interests which end up in endless discussion without any result. In 2000, we had deadlines and needed to deliver. We couldn’t wait any longer until everybody was in agreement that why we just developed our own schema back then and left OTA aside. Later our schema somehow became the standard – at least for air. A decade later our schema was introduced by IATA as NDC. But now the same problem happens again: everybody wants to add his own preference and all of a sudden an air shopping message becomes so huge, that it cannot be easily handled anymore. Thus developer refrain from using it.

In order to become successful as an industry compromises are important – even if it means that not all of your preferences are reflected. You also need to be fast if a demand for a certain feature is there to get it standardized. On the other hand something which was developed need to remain stable and unchanged for a good while. We haven’t changed our schema in 5 years until recently when we introduced ancillary and our clients love us for that because they don’t have to change anything on their end all the time. Of course, we have still made progress and introduced new optional fields/ features but nothing that required change for our regular clients. In less time IATA introduced NDC and overhauled it completely several times. This makes it very complicated for people to trust that their investment is safe and has future if you constantly have to change and adapt.

Some airlines can’t wait or don’t want to wait and they move ahead. It is hard enough for them to get their internal folks behind an idea, so they just skip the idea of a combined industry approach. It is not the best, but it is the nature of standardization. Mobile GSM probably was one of the single most successful standards ever created by mankind, but it was very late adapted in the US. So there is not ideal solution here, bottom line is you need to work with what is out there and see if a compromise is possible. Once a standard or pseudo-standard has been established, you need to remain it constant for a significant amount of time (target 5 years). We are not in consumer electronics here, we are talking about big, heavy systems that need adjustments. Not to forget ever rising security demands which add another dimension of complexity.

Ai: In the context of recent developments such as Siemens and Volkswagen now deciding to book flights directly via LHG airlines’ platform, how the distribution status quo is being challenged?

Michael Strauss:  It is not for everybody and it appears it wasn’t easy and took over a year to complete. It was politically strange that Amadeus Cytric bypasses Amadeus GDS to hook into Amadeus Altea. TMC involvement (created booking can be accessed by the TMC) turned out to be a huge challenge – so yes, there is some flexibility there, but in the long run I still believe at some point GDSs will catch up with more flexibility, new heads will revoke existing decisions and enter into new contracts and we will be back to the old environment just with a little more airline flexibility.

Ai: How are TMCs getting affected with the decision to look beyond GDS distribution?

Michael Strauss: TMCs definitely have their values and are much needed but with their dedication to GDS mainly due to the overwrites coming in from GDSs this is dangerous game that if airlines decide to bypass the chain as they are unable to change the rules with any of the player, TMCs might be degraded to value added services (Security, complicated bookings, remote areas, etc.) but much less the brick and mortar stuff which in the past have paid the bills. See next question about that TMCs need to reinvent themselves.

Ai: Going forward, how do you expect initiatives from airlines to change the world of corporate travel management?

Michael Strauss: I don’t see huge change, eventually all will stay the same – once the airlines are happy to have a little more influence to distribution and be able to position their product correctly. There will probably be some island direct connects but only in isolated areas - such as domestic Germany between huge players there. Even in Germany the prosperity of Germany is not built on Siemens, Lufthansa or Volkswagen, it is SME and no solution is anywhere near for SME. It is also interesting to note that LH only transports 4% passengers of compared to Deutsche Bahn. It is a good threat the airlines have against the oligopoly of the GDSs and for that reason they will keep this initiative alive, but in the end there will be arrangements.

Way more interesting to me is what TMCs will do, because they should really reinvent themselves and be more than just a call center using technology from everybody else. Otherwise I can see a shift that corporations will bring the technology and expertise inhouse and drastically reduce their agents. I see this happening with one of our financial clients. Don’t forget that soon there will be pretty smart AI agents.


Where is industry headed with developments such as NDC? Hear from senior industry executives at the upcoming 7th Mega Event Worldwide 2016, The Event for Loyalty, Ancillary & Merchandising & Co-Brands, to be held in Toronto, Canada. (25 -26 October, 2016).

Twitter hashtag: #MegaEvent16

Follow Ai on Twitter: @Ai_Connects_Us