Predictive analytics is emerging as a critical tool for gaining control over the decisions made on a daily basis and overall for organizations to successfully meet their business goals. It helps in prioritizing projects and investment in resources with expected revenue and profitability impacts.
Companies gather massive amounts of data on a daily basis. The key lies in making the most of this data in the form of insights and finding ways to use these insights in enhancing decision making. Predictive analytic solutions help in acting on such data for evidence-based decision making.
Travel companies, too, acknowledge the significance of predictive analytics.
Carnival Cruise Lines’ VP of Customer Marketing Rob Borden says predictive analytics are an important and ever-increasing factor in the company’s resource allocation and prioritization decisions.
Providing an insight into the usage of predictive analytics to mine and analyze the data gathered from web traffic and bookings on a website, Borden said, “We have just begun to scratch the surface of what is possible here. In what may be the most basic of ‘predictive analytics,’ we’re using site search and browsing history to send abandoned booking communications that are delivering outstanding results. Concurrently, we’re getting smarter at predicting the type of cross-sell and up-sell products to introduce between the time of initial booking and departure.”
Edward Nevraumont, Senior Director – Customer Loyalty, Expedia, in a recent interview with EyeforTravel’s Ritesh Gupta about the utility of predictive analytics as part of the group’s advanced data management strategies, mentioned that one of the ways in which the group is involved in the same is for the analysis of the profitability of different types of customers.
Explaining the structure, Nevraumont said: “At Expedia Joe Megibow (VP, Global Analytics and Optimization at Expedia.com ) runs a very strong analytics program that is built on a strong foundation of data warehousing – I believe we will have several hundreds of terabytes managed by FY11, most of this in a clustered DB2 instance that is accessed throughout the company. We literally run our business by these numbers and have a team of about 30 to 40 analysts that maintain dashboards, write custom SQL queries, and do all sorts of structured ad hoc analyses.”
Expedia also has a smaller team that builds predictive models that complement these types of analyses.
Задържане на клиенти
One of the most useful applications of business intelligence is developing effective customer retention solutions and reducing costs by identifying the greatest number of customers likely to churn within a small percentage of your customer base. Predictive model tells you which new customers are likely to return and which are probably one-timers.
A group like Expedia studies this closely and has several predictive models based on RFM (recency, frequency, and monetary) statistics. Expedia has found that these measures to be “very good” predictors of the propensity of a customer to shop with it in the future. The team provides goals within its organisation to make sure these measures are what it wants them to be.
From Carnival Cruise Lines’ perspective, Borden pointed out that the cruise business has a huge volume of first time guests.
“In some cases, greater than 80 percent of all guests are first timers, or rookies as we call them, at Carnival. Being able to more accurately predict which guests are likely to return is critically important so that we may focus our marketing efforts and dollars accordingly. Given the infrequency of the cruise purchase, we’re just now seeing the early results of our efforts to predict which guests will re-cruise with us,” he said.
Travel companies are trying to deepen their understanding of the current and changing purchasing patterns of their main group of customers.
The focus is not only on understanding who buys what, but why they buy, when they buy, and how much they buy. From these insights, companies deliver increased value to their customers.
Behavior analytics helps in segment and understand certain customer groups and then anticipate new customers’ needs.
“We use guest behavior in a number of ways. First, for past guests, our ability to speak to you based upon your expected lifecycle is showing great results (dreaming/researching, shopping, purchasing, not purchasing, win back). Also, for past guests, we develop a profile based upon their past onboard spending behavior to develop offers and predict a guest’s lifetime value,” said Borden.
He added, “For prospective guests, we actually ask a few key questions in order to provide relevant content and offers to them and to help overcome a number of common barriers to cruising.”
According to industry experts, in recent years there has been a movement from behavior being purely financial to analysis of behavior engagement across transactional metrics, responses, engagements, referrals, clicks, and in-store purchases. This is the most critical element of brand metrics. It is the one thing marketers can affect in analytics. The rest has to do with understanding who the customer is and their motivation for purchasing, which is difficult to control.
Borden agrees with the same and says: “We’re increasingly laser-focused on measuring our campaign response, guest engagement, referrals, and other means of understanding marketing effectiveness. It’s a wonderful time to be a marketer, as there are more metrics available than ever before to demonstrate the value of marketing investment to the skeptical.”
Customising customer experience
While it is imperative to customize customer experience and build a one-to-one bond, it is equally important to make sure one does the same without infringing on personal privacy and risk “turning off” customers.
“While this is a fine line, because of the unique nature of a cruise and the even more unique nature of the Carnival guest, we are able to deliver some truly memorable, one-to-one moments,” shared Borden.
He further added, “The Carnival guest is truly a ‘spirited’ individual and typically quite outgoing and gregarious. They eagerly share information and in combination with observed behavioral data, we provide our onboard crew with important information so that they may personalize and individual[ize] the guest experience in a number of ways. We do find that the tolerance for this kind of one-to-one experience is much greater onboard than off. Thus, we must use a greater level of care and discretion in our marketing and communications.”
Carnival Cruise Lines’ VP of Customer Marketing Rob Borden is scheduled to speak at the EyeforTravel’s Customer Centric Strategies in Travel conference, to be held in Atlanta next year (January 26-27, 2011).