NB: This article is written by Dr Kelly McGuire
Executive Director, Hospitality and Travel Global Practice, SAS Institute
Revenue management is taking on a more strategic role in the hospitality industry, with growing responsibilities and new opportunities to expand the discipline and its influence. In the midst of this evolution, one thing remains constant: the person who fills the revenue management role is the key to a successful revenue management program. This role requires an analytically minded leader, who can successfully work across the organization to drive results.
Revenue management has traditionally been a detail-oriented, analytical discipline obsessed with data and consumed with spreadsheets. The more analytical, the better. Even as few as 10 years ago, revenue managers were building their own revenue management systems in Excel, and running the pricing of large hotels entirely using spreadsheets. As the discipline has matured, the tools and technology have become more sophisticated and the role has expanded, the requirements to be successful in the job are continuing to evolve.
It is important to remember that analytic solutions are decision support tools. They may be configured by analysts, but the results are consumed by intelligent managers who have the experience to interpret the results and take the appropriate actions. Revenue management systems drive revenue because revenue managers can interpret the price and availability recommendations as part of a broader pricing strategy. The job of the revenue management system and the revenue manager are not the same. A hotel cannot simply hook up the revenue management recommendations to their selling system and walk away. However, a great revenue management system managed by a business-savvy revenue manager is a winning combination.
Recently, a revenue leader from a large hotel brand told me that one of the driving factors for their business analytics programs is to get better information into the hands of their senior executives faster. “Imagine how much more effective smart and charismatic leaders would be in a contract negotiation, internal strategy discussion or owners meeting if they had instant access to key performance metrics that can be sliced and diced to support whatever direction the conversation may go?,” she pointed out. “We have very experienced leadership, but I’m sure they could drive more revenue with access to better information, faster.” It’s not that the information itself doesn’t exist. There are always standard sets of reports available. The problem is that if a question is asked beyond the scope of the existing report, whether it requires additional data or a different view of the existing data, the additional data access and ad hoc analysis is usually time-consuming, not instantaneous. The advantage that faster and more flexible analytics provide is in the speed of access to the information, and the flexibility of the output. Delivering this is beyond the capability of most organizations today.
The primary purpose of decision support systems are to augment the existing experience and acumen of a top-performing executive, providing information for them to better interpret a situation, reinforce a point, convince an investor or make a key business decision. The right decision support tools, backed by credible data and advanced analytics are important, but it the right person in the role of interpreter and decision maker is crucial.
This is why I argue that we are at an inflection point in hotel revenue management. We are moving through the chain of analytic maturity, and we are getting to the point where we will need a different type of revenue manager to move forward and stay ahead. As the needs of the business change, the skills sets and competencies of the revenue manager must evolve as well. In the next section I will walk through the three stages of analytical decision making and describe how revenue managers must continue to evolve to keep organizations successful in today’s competitive environment. Descriptive: In the early days of revenue management, it was all revenue managers could do to access the data to develop and interpret historical reports that were built primarily in Excel. This was the descriptive phase. The organization could calculate that occupancy ran about 80% last month, or that 40% of reservations booked in the week before arrival. Past revenue was tracked to identify historical trends. Decisions based on this historical snapshot primarily involve reacting (i.e. putting out fires).
Accessing data from disparate systems challenged revenue management, as did the limitations of Excel. The majority of analyst time was spent in data preparation, with little time left for analysis or decision making. Creating these reports was time consuming and prone to mistakes, but it at least gave the organization visibility into patterns in the data. Rules of thumb, like threshold pricing (raising prices as occupancy increased) were put in place to help drive revenue. As more information got into the hands of decision makers, they were able to react faster, but were still only reacting to historical patterns. Still, revenue improvements were achieved by putting rules of thumb or operational procedures, like overbooking, in place based on the historical data. This also helped organizations evolve to the second phase of analytical decision making, the predictive phase, because they began to conceptualize the benefits of becoming more predictive, as well as demonstrate sufficient success to justify the investment in predictive technology.
In the next state of analytical evolution, revenue management began to deploy advanced analytic techniques that allowed it to identify trends and take advantage of opportunities. Hotels started to apply forecasting, predictive modeling and optimization algorithms to existing transactional data. The pioneers had to hire teams of forecasting and optimization modelers to create proprietary revenue management systems from scratch, but eventually vendors came on the scene and built analytical systems that could then be configured to a hotel company’s operating conditions. Not only did these systems provide access to advanced predictive analytics, but they also streamlined and automated the process of extracting data from the source systems, transforming it into clean, analytics-ready observations and loading the data into the revenue management system. They also automated the process of updating the selling systems with pricing recommendations. This freed up a good deal of the analysts time for interpreting and deploying pricing recommendations. As the discipline and the systems evolved and achieved success, revenue management systems became main stream, and turned into practically a “must have” solution for hotels.
The predictive analytics models in revenue management systems produce results like: occupancy will be 80% next month, the optimal price for a suite is $409, or demand is expected to trend down for the next several months. Organizations then prepare themselves to manage through these expected conditions. They are able to be more proactive in their approach, pricing to take advantage of peak periods or discounting to drive demand. This predictive decision making means that hotels set the “right” price in anticipation of taking advantage of demand patters as opposed to reacting to increased occupancy by raising price after occupancy has already reached a certain level.
Revenue management, in general, is at this predictive phase today. As a discipline, we understand the value of proactive decision making using advanced, predictive analytics, and have set up our organizations to understand and implement the revenue management system recommendations. This is a major achievement for the discipline, and for hotels. In fact, revenue management is leading the entire hospitality organization towards predictive, proactive, decision making as it has demonstrated success in driving profitable revenue. This success, and the mainstream application of revenue management in the hotel industry, is driving revenue management towards the next, necessary, evolution in analytic capability.
The final stage of analytic evolution is to change the mindset of decision making from, “what happened” and “what will happen” to “what are we going to do about it?” Prescriptive decision making relies on analytic techniques like optimization, which provides the best possible answer given all business constraints, or simulation, a “what-if” technique in which a complex scenario with multiple moving parts is modeled so that options can be tested to evaluate the impact on key outcomes.