In a business class of its own

Expedia travels to new heights using SAS® Business Analytics

In the highly commoditized world of e-commerce, which has historically been focused on high transaction volume versus better customer insight, Expedia — a pioneer in offering online hotel, plane and car reservations — exemplifies the definition of a modern-day, data-driven organization.  Expedia is using SAS Business Analytics to optimize online customer experiences, as well increase the lifetime value of each customer.

By striving to better understand its customers' needs, Expedia is using analytics to mine terabytes of business and online transactional data to discover innovative ways of increasing customer satisfaction and loyalty, while increasing annual gross bookings and revenue.

Competing on analytics
"We have organized ourselves to be a data-driven organization competing on analytics," says Joe Megibow, Vice President and General Manager, Expedia US, who recently accepted a Web Analytics Association award for practitioner of the year. "Almost all of the heavy lifting we do to intelligently use data is done with SAS."

After a significant project by Megibow's team to centralize the company's data into a global data warehouse and apply standardized data definitions, Expedia is effectively using analytics to: understand what marketing promotional channels influence and drive revenue conversions on the site; optimize its marketing spend by channel; increase customer lifetime value; and improve the overall experience of customers while on the site. The latter, of which, recently helped the company avoid millions of dollars in reservation losses.

Creative analysis
Driving business results with innovative uses of analytics is Expedia's specialty. In one example, Megibow describes how a member of his team suggested using the Cox regression model — used primarily in the pharmaceutical industry for drug testing trials — to cut through the clutter of customer touch point data and positively attribute the origin, or marketing influence, of customer transactions.

"Rather than doing a more traditional regression or clustering analysis, we tried this model out and had phenomenal success," explains Megibow. "We employed a model that looked at all touches across all customer sections to figure out what really influenced a transaction with us. We discovered that we weren't even close with our prior models. What this means is that we spent increased levels to drive incremental transactions, which, using the old model, would have looked inefficient but was actually more profitable. We would have spent too little and made less money — we didn't know that we were leaving money on the table.

"SAS inspires creative analysis," he continues. "We may know what the problems are or know where we've done something wrong, but SAS helps us figure out how to do things better. We actually have a backlog of opportunities from research that we've done. We've been able to capitalize on SAS' experience in solving so many different problems in so many industries. This is an unsung asset of SAS that helps creative thinkers apply solutions in new ways to solve challenging problems."

Customers for life
According to Megibow, one key area of focus for Expedia has been understanding and improving customer lifetime value. This can be a particularly difficult task given that many customers who reserve through Expedia do not always create an account, which means Megibow's team had to connect the dots to better understand the behavior of its customer base.

"Our industry has some commodity aspects, which can make it tempting to focus more on transactions than customers; increasing lifetime value requires a long history of customers, not transactions," he explains. "We spent significant time building data sets, clusters and predictive models to determine which customers are most likely to buy. We can determine how much we are willing to spend to acquire a customer on the first transaction, and which types of transactions or campaigns have the highest probability of driving a second or third transaction. Customer lifetime value analysis also demonstrates that the cost to drive future transactions declines among loyal customers. We can spend more to acquire customers early on, but the overall cost for transactions declines as we come up with programs that build loyalty."

On the subject of loyalty, Expedia recently launched a loyalty program with partners, which was enhanced based on the customer lifetime analysis project.

"A good rewards program isn't about just giving money back to good customers," explains Megibow. "A good rewards program is about understanding who customers are so that you can encourage the right purchasing behavior and improve their overall experience with your brand. For example, if I have a loyal shopper who books lots of flights but never buys a hotel, we want to offer them a good discount to try out our hotels or packages. The program will allow us to do that and it's a win for our customers. We can give them something we wouldn't otherwise just give away to everyone."

Relevant offers, increased conversions
With so many flight options, hotel choices and car rental offers available today — for a multitude of destinations around the world — analytics is also helping Expedia to generate relevant offers when customers come to the site to reserve, for example, hotel accommodations.

"We built and trained a utility model, using real-time pricing data from hoteliers and historical customer data, to predict what would be most relevant to consumers. We've been doing it for a couple of years and have seen multiple percentage point increases in conversions," Megibow says.

Improved usability saves millions
To improve the online customer experience, as well as avoid lost revenue due to site usability glitches, Expedia also employs analytics to help developers discover and fix issues that might be impeding customer satisfaction and new revenue opportunities. In one such case, the identification of a usability glitch saved the company from losing millions in annual revenue.

“Our standard billing area had an optional field asking for company name right above the required billing address information,” recounts Megibow. “We discovered that some customers were confused by our form and entered the company name of the credit card issuer, and then went on to enter the address of the credit card company.  Once we discovered this we merely deleted the field and saw a massive conversion change — we saved about a million dollars a month in lost business.

"We are now able to fund new projects as a result of all the analysis we've done, and make the informed decisions we've always wanted to make," he concludes.

The results illustrated in this article are specific to the particular situations, business models, data input, and computing environments described herein. Each SAS customer’s experience is unique based on business and technical variables and all statements must be considered non-typical. Actual savings, results, and performance characteristics will vary depending on individual customer configurations and conditions. SAS does not guarantee or represent that every customer will achieve similar results. The only warranties for SAS products and services are those that are set forth in the express warranty statements in the written agreement for such products and services. Nothing herein should be construed as constituting an additional warranty. Customers have shared their successes with SAS as part of an agreed-upon contractual exchange or project success summarization following a successful implementation of SAS software. Brand and product names are trademarks of their respective companies.

Copyright © SAS Institute Inc. All Rights Reserved.