First Published on 5th February, 2018
Ai Editorial: Technology continues to evolve and airlines have to come to grips how the same is facilitating sublime passenger experiences, especially via a digital asset, writes Ai’s Ritesh Gupta
Airline-owned digital assets need to keep pace with the latest advancements, varying from relying on artificial intelligence to automate personalization to contextual delivery of content to embracing agility for product development.
Some of the emerging developments that airlines need to consider:
Flexibility over operations and deployments: “There is a lot of pressure on airlines to deliver new experiences to passengers,” acknowledges Kevin O’Shaughnessy, CEO and co-founder, Indigo.gt. So there is a constant need to evaluate mobile interfaces, web interfaces, and even service touchpoints. In this context, there is reference to “microservices”, termed as an advancement of the service-oriented architecture design. In this case there is one complicated application that is divided into smaller bits that are simpler to develop and test against. Each microservice has a self-contained set of functional duties and easier to support. Thus airlines or travel e-commerce end up having more flexibility over operations and deployments.
When it comes to changing the way purchasing flow or the check-in process works, these tend to be relatively big IT projects. But when it comes to aspects like microservices, O’Shaughnessy contemplates the possibility of appointing a project manner, responsible for “getting something done quickly and get out of the door in an agile way”. This is an interesting approach to product development for the airline project manager. So consider the option of “microservices for your own project”, in addition to big IT projects and other approaches to IT set up, he added. “In case of a quick (and a fairly smaller) project, one could possibly turn to microservices (so typically a case of a software development for one service, might be an API or plug into something else) and chances are that the airline IT team would already be using associated services. So it could be about finding a way to running an experiment quickly,” explained O’Shaughnessy. Of course, the big picture doesn’t change, but this methodology brings in agility to embracing new concepts – for instance, testing WeChat or Alipay payment options with the elite members of an airline. The role of application programming interface (APIs) in formulating flexible ecommerce architecture can’t be understated.
Talking of mobile or web interfaces, increasingly there is focus on decoupling the front-end user interface of the ecommerce storefront from the main commerce offering. Such methodology ensures quicker crafting of experiences that upkeep with various platforms and devices or chip in with new travel products.
As for moving infrastructure into the cloud, it is time to leverage today’s technology at a much lower cost, and avail the benefit of scaling it up. Airlines consider factors such as such as security (the role of technologies such as encryption and tokenisation comes into the picture) as well as connecting legacy applications to the cloud at enterprise scale.
Experience optimization: Capabilities of the content management system available today is one area that needs to be factored in, first with the proliferation of different devices that didn’t use HTML and also with the Internet of Things. But the delivery of content as per the device being used is just one aspect. E-commerce players also need to able to create and edit content in context, and pave way for personalized experiences. The customer experience ends up being decoupled as well. This would limit the ability to personalize the overall experience.
As highlighted in a previous article, there is a balancing act that needs to be managed, when there is consideration of headless or coupled. Of course, with a headless content management system, one can create and store content that is device agnostic. On the contrary, as Sitecore points out, “what-you-see-is-what-you-get” capabilities that let users create and edit content in context aren’t there with headless. Since personalisation calls for gathering and study of user interaction data from the front-end in order to optimise personalised experiences from the back-end, it requires loads of custom integration work in a headless architecture since the front-end and the back-end are disconnected. Content management, be it for going for an architecture that supports delivery of content for emerging technologies and all devices, or adopting personalisation rules that tailor a site content based on visitors’ profiles is one key area that is demanding action in a swift manner. So when it comes to a headless content management system, Sitecore asserts that such architecture should be used by “digitally mature” entities that are capable of managing “customer experiences in context of how users interact with your brand”. Further adds, it should work for those “whose digital properties are personalized, who regularly test and optimize those experiences, and whose organizations are set up to be customer-centric.” Then only one should expect to manage the balance between contextualized digital experiences and standout app user interfaces.
Counting on machine learning: The application of machine learning in managing e-commerce fraud is coming to the core. The issue of fraud or account takeover can also hamper the user experience. As much as consumers experiment and embrace new forms of payment options, each new technological development introduces new avenues for fraud, meaning detection and prevention efforts need to be just as agile. Many fraud prevention methods introduce dilemmas between maximising revenue and minimising fraud – e.g. with more rules, implementation of 2FA or multifactor authentication fraud rates can be lowered, yet more genuine customers will be blocked; on the other hand, with less rules and lax authentication to maximize revenue, merchants will be more vulnerable to fraud attacks.
Companies can defend themselves adequately by using a tool like machine learning, and at the same time there needs to be reliance on rules and the human component (intervention and feedback) as well. The power of machine learning is still in the supervised state as of today. Typically, the supervised machine learning technique focuses on a cycle of training, predicting, and acting stages. Of course, airlines need to look around data quality and data volume in terms of how clean that is to capitalize on machine learning. Once a virtuous data pipeline is in place, it can be built upon with machine learning models, with rules, and create tools to help the team analyze problem areas such as account takeover.
Hear from experts about optimising digital assets at the upcoming Ancillary Merchandising Conference, to be held in Edinburgh, Scotland this year (9-11 April, 2018).
For more info, click here
For Ai’s 2018 Events, check - www.aieventdates.com
Follow Ai on Twitter: @Ai_Connects_Us