14 July 2017
Using Data and Insights
How to drive ancillary sales
At the recent Future Travel Experience conference in Dublin, Boxever chaired a discussion on “Leveraging AI to drive Ancillary Revenues” as part of the Co-Creation Forum. Facilitating the debate was Paul Murrell our Client Solutions Director who has over ten years of direct airline experience in using data and data science to grow ancillary sales.
Whilst the conversation covered a lot of ground, inevitably it started with data and specifically, what are the most relevant data sets that an airline can leverage to put relevant and personalised ancillary offers in front of its customers.
Firstly, we discussed what we thought were the main customer / booking attributes we felt influenced ancillary sales. The ones that were deemed most important were Passenger Type (business vs leisure), Passenger Mix (Family, Couple, Single), Timeliness in terms of the booking window (6 months vs same week for example), Timeliness again but in terms of the point in time when the booking is made compared to the date of travel (6 months out vs same week) and then Channel (inbound vs outbound and digital vs face-to-face).
Another attribute we deemed worthy of consideration was the booking class and, specifically, the table considered whether they would use it to flex the online booking flow; if a customer was browsing for a fully flex ticket with a high margin, would we simplify the booking flow, bypassing the ancillary sales page and fastrack them straight to payment?
Conversely, if they were browsing a low-margin restricted fare, would we almost gamify the booking flow to make it difficult to purchase unless they did purchase an ancillary? The table didn’t reach a consensus on this one but it made for an interesting debate!
About the relative merits of historical and in-session data
The conversation then focussed on the relative merits of historical and in-session data. It was acknowledged that historical data; that which reflects what we have previously observed, has a subtly different role to play than in-session; what we are observing right now.
The idea put forward was that in-session data was highly influential in deciding when to place an offer in front of a customer. A combination of in-session and historical data drives the decision about which offer and then historical data (and propensity models) could drive the type of ancillary (e.g. which star rating when it comes to a hotel offer).
Take the example of a customer making a midweek, short-haul, same-day return booking online that we have observed them making regularly. Here, using simple rule-based decisioning, we can assume that this is a business trip. It would make sense then when it comes to promoting ancillary offers within the booking flow to upweight Paid for Seating (they can maximise their time by departing the aircraft quickly), downweight Baggage (they are unlikely to have any) and downweight an overnight hotel in destination (they are returning later that same day).
Using previously observed behaviour to further tune recommendations
However, if we can identify from the customer’s profile that their home address is greater than fifty miles from the airport then we could upweight a hotel recommendation for the evening before the flight near the airport, especially if we have observed this behaviour previously. We can then use previously observed behaviour or a propensity model to further tune the recommendation to a particular star rating of hotel or even a specific hotel.
What this example demonstrates is that to be able to put the right offer in front of the right customer at the right time then we need to bring the online, in-session data and the offline order, product and customer data together in the same place alongside any analytical models. It all needs to be accessible to a decisioning engine or ‘brain’ where the business team can build out those rules that bring to life the great ideas they have and see them realised in a truly omni-channel manner.
This is just one, relatively straightforward, example of using data and insight to drive ancillary sales; every airline will have many more great ideas, they are just being held back by the capability to deliver them. It was this observation that led us, at Boxever, to build the Customer Intelligence Cloud, a platform for marketers that connects all of your customer, product and operational data, putting your customer at the center of your business and enabling true 1:1 personalization.
Using artificial intelligence (A.I.), it acts as the “brain” within your customer-tech ecosystem, taking in all data about the customer, deciding what should happen next and executing that action through the most appropriate channels – in real time, as it happens.
As we move forward in this ‘age of the consumer’, decisioning can play a critical role in helping organisations meet – and exceed – expectations by unlocking personalisation and driving market-leading CX. But it’s clear the benefits don’t stop there. Companies can turn decisioning to their advantage, reinvigorating the entire organisation with a more agile and – ultimately – more profitable way of doing business. Seen in this light, the prize is simply too big to ignore.
Director of Partnerships
Paul is Director of Partnerships at Boxever – a market-leading personalisation platform that uses data and AI to make every customer interaction smarter. Boxever is recognised by Gartner as a leading player in personalisation and ranked by Forbes alongside Google, Apple and Amazon as one of the most powerful examples of AI in use today.