For many organizations across the marketplace, the promise of analytics has yet to be fully realized. And in an industry still trying to evolve from mainframe computers and legacy systems, transitioning to an analytics-led business is all the more complicated.


Travel and transportation companies are working to better leverage sales and marketing analytics to get closer to the customer, and many of the hurdles that they face aren’t atypical. With the help of the Economist Intelligence Unit, ZS recently completed a cross-industry study to assess the state of analytics implementation and integration, and found that companies’ results thus far have been underwhelming. Among 448 respondents, 70% rate sales and marketing analytics as “very” or “extremely” important to their competitive advantage, but just 2% report that they’ve managed to generate a “broad, positive impact” from their analytics investments thus far.


Glenn Hollister, a principal at ZS and leader of the firm’s travel and transportation practice, offers his perspective on the state of sales and marketing analytics in his corner of the marketplace, and what’s on the itinerary as organizations begin to evolve.



A: I think, in some ways, the results are almost too optimistic. Some of the shipping and logistics companies get a lot of value out of analytics, but that’s primarily on the operational side to generate efficiencies through things like route optimization. And some of the newer entrants and Web-based companies are getting good at analytics. But most of the well-known, established categories—airlines, rental cars, hotels, travel agencies—really struggle to get much value out of sales and marketing analytics.

A: One big area is talent. A lot of these companies typically don’t have a large staff of people who are very analytical. Some large airlines are starting to develop analytics centers, and once those are up and running, they’ll have pretty good capacity. But if you look at where they were six months or a year ago, they had a handful of people at most.


Also, while they’re smart to consider the talent side of the equation, I think that some companies may struggle to build it up internally. The compensation for people who are really talented analytically is quite high, and the kind of workplace environment where they thrive is quite different than travel companies typically offer. It’s a very different culture, with a different rewards system and other variables. Because of that, some companies may build these teams internally and find a lot of churn, as their best analytics employees get poached by other companies, either in travel or in other industries.


The other barrier, and one that's even harder to surmount, is that the IT systems in the industry tend to be quite old and very fragmented. A lot of the core systems in the travel industry still run on mainframe computers that were first programmed 20 to 30 years ago, so in a lot of cases, you have data fragmented across multiple systems. That requires a huge effort just to pull the data out of these legacy systems and get it cleaned up and aligned to the point where you can even start to make sense of it.


This is especially true for master data, a comprehensive database that includes all customers and all assets so that transactional data can get tied to that. If I want to know, say, how many corporate customers stayed at a particular hotel or what percentage of passengers on a specific flight met precise criteria, it’s very hard for travel companies to answer those questions now. They lack good, clean master data and a reliable system. Companies are replacing some of the old, core IT systems, but until they build really reliable master data systems, they’re always going to struggle with analytics.


Q: The study also found that even when companies get the analytics right, they sometimes struggle to translate those insights in ways that directly improve the customer experience. Does that happen among travel companies?

A: Definitely. If the analytics suggest doing something different, you’ll get people who say, “Well, our systems don’t support that.” And operationally, it’s so hard to actually make a change.


Think about the rental car industry: They can improve satisfaction by making sure that they get each customer the car that he or she wants, based on their past rental history and preferences. Maybe they prefer a certain size or make, or they really need Bluetooth in the car. It’s actually not that hard to track that information, but getting that car to that customer is another matter.


At a big airport operation, they’ll have dozens of cars, but they’re all parked nose to tail, so only a small number of those are readily accessible. The guy who runs back and forth to pull cars out might have a sheet of paper—or maybe a tablet—showing that day’s rentals and their ideal cars, but all he can do is make the best match based on which cars he can get to. And sometimes it’s more like which cars he’s standing closest to.


You can do great analytics, mapping out the entire fleet and customer base, but putting that into practice to change the customer experience is extremely hard. But people recognize the potential. There are a number of senior executives in the industry who understand the power of analytics, and the industry overall is used to operating with data and making fact-based decisions. The challenge is understanding what’s required to actually execute on that promise and put it into practice operationally.

A: At most, they can do basic stuff like descriptive analytics, not predictive or prescriptive analytics, and they’re slowly starting to get better at pricing and product design. For example, some airlines are starting to roll out branded fare packages with varying bundles of service. Instead of simply buying a ticket, you can pay a bit more and board early, or you get an additional checked bag at no extra charge, free drinks on the flight or additional miles. When you book online, you’ll see all of these options, which have different prices.


That involved an analysis of what kinds of features customers want and how much they’re willing to pay for them, but it’s typically a one-time run. The fares vary depending on a customer’s frequent flier status, but those changes are predetermined.

A: What everyone in travel wants to do is dynamically create and price a package while the customer is shopping in response to market conditions and the customer’s browsing or purchasing history. And prices should change in ways that ultimately improve profitability, but that kind of real-time, transactional analytics is still very much in the realm of theory. There is a very small number of airlines that are varying pricing for packages based on analytics, but none are varying the packages, themselves.