The third-party wars are raging in Europe with rate parity’s demise pending only a few more strokes of the regulatory pen. With pushback on last room availability by many hotel companies, inventory control may soon be back in the hands of the hotels. Also crumbling is the long embraced theory of the billboard effect. Expounded by the OTAs for the last 10 years, the billboard effect claims to provide “free” merchandising, leading to even more bookings direct on a hotel’s own website.

Ultimately, even in this changing environment, hotels in today’s digital landscape have to constantly balance the two options they have to acquire and retain customers: direct initiatives and those deployed through third parties. There are costs associated with every single acquisition technique. While many channels can bring value, none bring unlimited value or have infinite volume. Tradeoffs and decisions have to be made by hotel revenue strategists.

How do hotel management teams navigate these decisions to achieve an optimal channel mix? How do you decide how much to spend in total and how to allocate that budget by channel?

The Online Travel Shopping Path
A study conducted by the University of Maryland’s Center for Service Excellence* examining consumer behavior for online travel shopping will be released as part of the updated Distribution Channel Analysis report, and is due out the first quarter of 2016. A review of the so-called “clickstream” indicates which stops consumers make on the path to a hotel booking. This type of analysis is called attribution modeling and gives an objective view of consumer behavior when booking hotels, so that hoteliers can accurately figure out how to attribute credit for hotel bookings. Knowing what truly influences the bookings for any given hotel will help revenue strategists to decide where to invest limited customer acquisition funds. The data set in the University of Maryland study represents close to 4,000 hotel bookings spread across many brands, all based in the United States. This will help shed light on a much discussed but little understood area of the digital travel industry.

*The Online Travel Shopping study is produced by the University of Maryland Smith School of Business Center for Service Excellence and sponsored by the AH&LA Consumer Innovation Forum.

What Actually Triggers an Online Hotel Booking?
The billboard effect purports to give a simple explanation for a complex consumer behavior. While we would all love to believe there is a silver bullet that explains what delivers bookings to a hotel, unfortunately, the reality is far more complicated. The upcoming 2016 Online Travel Consumer Behavior study attempts to address this complex shopping process and ascertain which variables can be used to determine the degree of influence ascribed to each stop in the hotel booking shopping path. Does this change based on average rate? Based on hotel brand? Does the length of time on a site indicate its degree of influence on the ultimate booking? Is there a sequence of stops that creates a pattern that repeatedly culminates in a booking? Before jumping to the unlikely conclusion there will be one silver bullet in the noisy world of hotel shopping and buying, it may be more realistic and productive for a marketer to understand the many stops made by travel shoppers and test investments in multiple points along the way.

Is the Billboard Effect Dead?
One of the early results of the 2016 Online Travel Shopping study indicates that consumer behavior has changed dramatically over the last five years. Much of the data from 2014 hotel bookings indicates behavior that will not come as a surprise. OTA booking volumes have grown steadily since 2001 – a cursory review of SEC filings and annual reports for Expedia and Priceline make it clear that the majority of the OTA profits are driven by hotels and that these profits have not stopped growing since their entry in the marketplace. The data indicates that not only has there has been a significant rise in the use of OTAs for booking, but also that a large majority of consumers starting their shopping process on an OTA do not return to book on a hotel brand.com website. Whether or not this behavior existed in the past, the data in the upcoming study shows that consumers do not typically use the OTA to gather information and then return to the hotel branded site.

Earlier Studies—Some History on the Billboard Effect
In the past, there have been two studies on the subject of hotel booking attribution. The first study attempting to prove that OTAs triggered bookings on hotels’ own websites was published in 2009. This study represented data from one hotel management company that showed results from four hotels turning Expedia on and off for a period of three months. This methodology was limited in scope and did not indicate what other factors were at play in terms of availability or demand in each destination. It did not indicate what the competitive forces were for those four hotels. It also did not indicate what promotions were in play by the brands of those four hotels and what was being offered through Expedia or if these hotels had other promotions being run concurrently on other OTA sites. The lack of controls, along with the fact that there are only four hotels tested, puts into question the conclusion that credited Expedia as the influencing factor in 7 percent to 26 percent incremental brand.com bookings. It’s clear that OTA participation cannot be seen as the only variable and the researcher described the study as a “pseudo-experiment.”

The second study, undertaken in 2011, was from a similar data set that is being used for the upcoming 2016 Online Travel Shopping study. It was sourced from comScore panel data and examined bookings in the summer months for 2008-10. However, its analysis is very different. The data set consisted of 1,700 bookings made for IHG-branded hotels. Besides the issue of the study being limited to one hotel chain, the interesting points lay in what was not reported or answered in the original study.  One interesting fact (not reported) from the original clickstream data in the 2011 study showed that seven to 10 travel sites were visited prior to a hotel booking. One is inclined to ask what discriminates between those seven to 10 sites to indicate which one(s) actually triggered the booking. Who can say if it is website stop No. 1, No. 5 or No. 7, or in all likelihood, some combination?

It isn’t just a matter of which sites were visited, but the duration of each visit and what patterns occurred in the 1,700 bookings to indicate that a particular sequence may lead to a booking. The original study assumes that only one of those seven to 10 websites visited actually triggered the booking and implies the other websites have no effect. Concluding that an arbitrary number of stops at Expedia must mean that Expedia triggered the booking would seem dubious. Upon closer examination of the data set*, it was revealed that there were more stops at competitive hotel companies and at airline sites than at all OTA sites combined (another fact not reported in the results of the 2011 study). Some expressed concern that because Expedia sponsored the original study, it might introduce some bias in favoring that particular site as having undue influence in the booking process.

So, does that mean that because there are more consumers stopping at Hilton.com or Marriott.com (or at United Airlines or American Airlines), that these stops are “the” trigger for an IHG booking?  If so, this assumption might lead the IHG brand marketers to choose to advertise on Hilton’s or Marriott’s website, but this may be an inappropriate conclusion. One could also jump to the conclusion that advertising with the airlines could be most effective. However, most likely, as there are many stops on the path to a hotel booking, the credit for the booking may be more difficult to assign than simply adding together the one website with the largest number of visits and assuming it takes 100 percent of the credit for the booking while the other seven to 10 websites visited have no influence on the booking decision.

* The comScore data used in the original 2011 study was provided for analysis and published in the Distribution Channel Analysis study published in February 2012.

Another consideration for revenue strategists is commission structure. Back in the early days of hotel sales and marketing, before the GDS domination of travel agencies, hotels paid 10 percent commissions to agencies to source business, be it from existing corporate accounts or from discretionary leisure travelers. In the 1990s when the airlines faced bankruptcy and cut travel agents out of the equation, they turned to hotels to fill the void. Hotels have now paid a steady 10 percent to travel agents for 50 years. We have dynamic pricing and yet continue to offer static commissions. It may be time to re-examine this practice and consider paying dynamic commissions (i.e., higher commissions during need periods and lower commissions when the business will come without added effort by third parties).

At this stage, hotels should closely examine each channel of business and assess its value weighed against its cost. Hotels cannot afford to continue paying 15 percent to 25 percent (and rising) of what the consumer is paying to stay in their property. There are more and more channel vendors looking to build their businesses on the backs of hotels along with many more meta-search and social media sites offering direct bookings. The evolution of Google Hotel Finder, Facebook and TripAdvisor into instant booking platforms is enough to show where things are headed.

Is this cost sustainable? Some hoteliers call it the cost of doing business. This is easy to say during high demand periods but when the economy tanks, and those costs continue to rise, it will be harder to absorb. In fact, the costs may skyrocket as they did during the last recession when hotels got more desperate to secure limited demand. When the revenue drops, hotels may not be so willing or able to accept the idea that they have no control and have to pay third parties whatever they ask. After all, labor costs are a cost of doing business, but hotels proactively manage those costs; there is no passive acceptance of them.

It’s time to apply the same tight management of costs philosophy to customer acquisition costs that is applied to labor costs. Examining revenue net of acquisition costs and reviewing a monthly recap of each channel’s contribution to operating profit will yield a more accurate picture of a hotel’s revenue performance. The upcoming Distribution Channel Analysis study to be released in February 2016 by Kalibri Labs and AH&LA will provide extensive market-level benchmarks for many hotel types in most major markets on these new and important metrics. 

All channels can be valuable and even more so when hoteliers have control over their hotels’ inventory and rates. Finding the optimal channel mix and managing to that objective is the new mantra of the hotel revenue strategist. Long live GDS, long live the 800 number and long live brand.com.

Cindy Estis Green is the CEO and co-founder of Kalibri Labs. She is also the chief consultant to the AH&LA Consumer Innovation Forum, a research group examining distribution issues at an industry level representing hotel brands, owners and operators.