Cross-selling Products: Use Cases from Behavioural Data and Micro Frontends
By Naveen. S.R
As an Online Travel Agent, AirAsia the Asean super app collaborates with strategic partners to offer flight itineraries to over 3,000 destinations across 700 airlines.
The task for product engineers is to build the capability to cross-sell the super app’s products across various Lines of Business to be able to increase customer experience, hence increasing Click-Through Rate, Revenue, and Transactions.
The cross-sell platform (Xsell Platform) is built following the LEGO development strategy.
We have an in-house design system seeded called Phoenix, that houses a large number of sharable components across micro frontend apps and storefront mobile apps.
We have also created a widget called scriptlet built out of pure javascript and can be injected on any given mobile storefront experience or the web micro frontends.
The scriptlet widget provides options for choice of User Interface such as banners, carousels, etc.
The feed for the scriptlet comes from the Xsell Platform.
The Xsell Platform can provide the feed from the following data sources:
Here are a few user journeys on how we use these data sources powered by the Xsell Platform.
Multi Product Carousel
We built a carousel that displays to our customers our various products, from flights to hotels to the best travel bundle deals. For the launch of this, we went with static data powered by our data pipeline platform. Based on contextual information of the user — GEOID, Device, Culture, we served static data. The feeds can go through product boosting/suprress rules for LOB profitability.
With the capability of static data, we can test the waters, run A/B experiments to derive click-through rate before investing in more complex data feeds.
Contextual Feed On Search
To a user who browses AirAsia.com from Thailand and explores options on the “Book Flight” page, we promote a feed to cross-sell SNAP, our flight and hotel bundles at best price guaranteed.
This feed can further be utilized for bringing in data by user behavior over a dynamic feed. A user who has shown interest in certain destinations will then receive a feed to show the deals and best offers for those destinations.
Cross-Sell Hotels
With user insights of context, behavior, and transaction, we can power this over real-time data model and/or offline data model and derive cross-sell feed. The feed can further go through product boosting/suppressing rules to obtain special cross-sell rates.
A user on the confirmation page will get a relevant feed for the hotel based on the flight booking. We can bring back customers into other Line of Business funnels such as Hotels from white label platforms as well.
