To achieve greater returns on their machine-learning investments, aspirational travel marketers need to recognize and embrace even more analytical sophistication.
Leading marketers in the travel sector use machine learning (ML) not only to measurably improve business outcomes but to fundamentally redefine what those outcomes should be. Simply put, ML is helping travel marketers learn more about what outcomes should matter most. In particular, they’re deploying ML to learn much more about their customers much faster. ML is making travel marketing analytics more predictive.
Our recent global executive study of strategic measurement reveals that a majority of travel marketers embrace ML: 70% of marketing executives in travel have incentives or internal functional (marketing-specific) KPIs to encourage the adoption of ML technologies to drive marketing activities, and 79% report that they invest in new skills or training to make marketing more effective in its use of automation and ML. (See Figure 1.) Measured according to our KPI Alignment Index (discussed in more detail in the “Leading With Next-Generation Key Performance Indicators,” report1), a larger percentage of travel respondents are, relative to the overall sample, advanced users of KPIs. In sum, the travel industry is a quantitatively sophisticated sector making investments in training marketers in ML.
To drive greater returns on their ML investments, aspirational travel marketers need to recognize and embrace the following trends.
The Emergence of New KPIs
Using ML primarily to improve or optimize legacy KPIs can be a fool’s errand. Travel industry marketers — who often confront born-digital disruptors, increasing globalization, and tourists who’ve made “selfies” an integral part of the travel experience — recognize that new value creation demands new metrics to assess it. Using 100 times more data to wring greater efficiencies from existing KPIs isn’t good enough; serious marketers look at orders of magnitude to spark novel insights and test new business hypotheses. ML makes that kind of data exploration fast and feasible. Unsurprisingly, born-digital platform-oriented travel companies appear quick to use more data and more innovative analytic techniques to bring new KPIs to their operations.
As short-term rental company Airbnb set out to generate revenue and additional market penetration by converting guests into hosts, new KPIs surfaced. The campaigns to encourage travelers to rent their properties varied market to market. In some, the company used email marketing; in others it employed more traditional campaigns using radio. In all cases, these were a collaboration between two functions: marketing and operations. “It was a shared KPI,” explains former Airbnb CMO Jonathan Mildenhall, “because operations teams driving the business would come to us and say, for example, ‘In this market, in Seattle, we want to recruit hosts. And these efficiency targets, marketing, how might we meet those targets?’”
Realigning Marketing With Operations
The airline industry’s use of shared KPIs points to where other industries (and companies like Airbnb) are headed. At JetBlue Airways, Marty St. George explains his role as chief commercial officer with a marketing team reporting to him: “We never really had CMOs in this business. It’s always been more of a chief commercial officer or chief revenue officer, a job where you’re controlling all pieces of the business that are tied to revenue.” It’s no surprise that 71% of travel marketing executives have visibility into other C-suite leaders’ KPIs.
Siloed information can stymie marketers’ use of ML. The sales function, for instance, typically has a great deal of data that marketing can use to build its own ML algorithms. As ML use increases throughout the enterprise, marketers have the opportunity to align KPIs and ML capabilities. Partnerships between marketing and other internal groups become practicable — and, in some cases, vital — to provide access to data. St. George explains that JetBlue recently appointed a central data scientist to “make sure that all the databases talk to each other and, in fact, agree” to facilitate the design of seamless customer experiences.
Read Related Articles
More Opportunities for Personalization and Customization
Today’s savvy, selfie-taking travel consumer expects personalized assistance at every touchpoint. That’s why so many top-tier travel enterprises see ML as personalization engines that convert huge volumes of customer data into more bespoke insights and offerings. Preserving the customer loyalty of a “TrueBlue” frequent flyer is one thing; converting that loyalty into exploring vacation packages and travel bundles is quite another.
St. George notes, “We track your digital journey to make sure that we make it as easy as possible.” Understanding customer travel behavior has led to the creation of new products for the carrier, including JetBlue Vacations, which creates customized itineraries for consumers. Similarly, Airbnb’s Experiences platform, which enables hosts to share not only their homes but also their knowledge of local environments, generates new data sets as hosts and travelers alike use this service. Airbnb’s marketers can use its recommender systems to suggest personalized travel packages to users that integrate host offerings.
Machine Learning: Just the Ticket
Marketing executives in the travel sector are well aware that ML is a source of strategic benefit, and they have embraced the technology accordingly. Mildenhall observes that these days, everything in the travel sector — the consumers, the competitors, the technology itself — moves fast. To keep up, he says, “you need real-time data that’s driving real-time strategy that’s driving real-time market implementation.” ML technology, together with the appropriate functional KPIs and structure to support it, will most likely play an expanding role in the data-driven travel marketer’s toolkit.