It's a great time to be in revenue management! Demand is up, rates are raising, and the revenue management discipline is getting attention at the highest levels in the organization. After five years of economic growth, organizations are feeling comfortable enough to invest in people and resources to improve decision making. Revenue management has always been at the center of data and technology investment in hotels. In 2015, revenue management has the opportunity to guide organizations to the processes and investments that will make the most impact on revenue and profits.

There is no doubt that the role of the revenue manager has evolved significantly in the last decade. Revenue managers have taken on broader and more strategic responsibility in their organizations, including having to work more closely with sales and marketing, and getting more deeply involved in setting business strategy. In this article, I'll provide some tips in the area of data and technology for revenue management executives to prepare themselves and their organizations for success in 2015 and beyond.

It is very easy to get caught up in the day to day details of a very detailed job, but if you are not staying on top of the trends as they happen – and taking the time to think through their direct impact – you and your organization will be left behind. Hopefully the sections below will help you direct your efforts, so you know what you should be paying attention to as we approach 2015 and beyond.

2015 will be the year of (more) data. Data is the revenue manager's biggest asset and biggest limitation. Recently, the market has started talking about big data, and many argue that we are moving from an era of data scarcity to data abundance. There is certainly no LACK of new data sources offered to revenue managers, but abundance is creating new challenges. It is becoming increasingly difficult to sort out the value from the noise. Yet, at the same time, the inability to get to the right data or get the data at the right level of detail continues to limit (and frustrate) revenue managers.

Data storage is becoming cheaper, faster and more flexible. However, just because you CAN store the data doesn't mean you SHOULD. It takes resources, both technology and people, to manage data sets, large or not-so-large. It is also easy to get distracted by data, either getting lost in an analysis without producing any results, or by over-fitting a model with so much data that it becomes unable to predict the future. Don't become so consumed by the "big data" buzz that you are distracted from understanding the purpose and the value of the data. In fact, there is as much value in investing in going after what you "can't get to" today – more detailed demand by room type or cancellation information, for example - as there is in trying to get find the benefits in new data source.

Here is My Advice for Managing Through the Year of (More) Data

Get Back to Basics - With all the hype surrounding "big data" it is easy to get distracted by the next shiny (data) object that is put in front of you. It is easy to think that simply cramming every possible influence into the revenue management system will produce better results. Don't forget that the job of the revenue manager and the job of the revenue management system are not the same. The job of the revenue management system is to turn inputs about market conditions and demand patterns into day by day pricing recommendations for every product across the booking horizon. The job of the revenue manager is to evaluate those recommendations against the hotel's business strategy, apply their knowledge about the market and operating conditions and set a profitable pricing strategy.

Innovations in revenue management systems generally come from applying new techniques to or accessing new levels of detail in traditional demand data, as opposed to forcing new data into existing forecasting and optimization algorithms. Because of the complex and interrelated math involved, more is not necessarily better, for forecasts in particular. For the revenue manager, however, access to a broader set of data could certainly improve strategic decision making.

So, the advice here is to carefully evaluate any new data source, thinking through where and how it would be used in revenue management decision making, with an eye to the purpose of the revenue management system, and the job of the revenue manager.

Invest Smarter - There is a lot of interest in the market in modernizing revenue management, much of it focused on data and systems. Savvy players recognize that the revenue potential from the revenue management system depends heavily on how the data is collected by the transaction system and the recommendation format the transaction systems are able to take. Because of the critical nature of data inputs and recommendation formats, many companies are considering revising or replacing CRS and PMS systems in order to capture the right data to take full advantage of revenue management systems.

As you consider technology investments, carefully analyze your current selling systems and ask yourself whether you are losing revenue potential because you are not able to collect the right data, or implement the right recommendations. The smart investment may be to start there, instead of replacing your revenue management system, or to replace those systems as part of the initiative to upgrading your revenue management system, so that you can take full advantage of the new revenue management technology.

Be disciplined - As new data sources and or alternative data formats become available, develop a process for evaluating these sources, incorporating them into your analysis, and maintaining them over time. - Establishing a solid data governance process internally across the business, working with IT, of course, will ensure that you get the most out of your data investments.

Be sure you understand where the data comes from, what the "system of record" for that data should be, how often it is updated, what level of detail is available, how it fits with the rest of your database, who it might be useful to, what you would use it for (reporting? Analytics?), who has the ability to change or manipulate it, who can simply view it, and who should not be able to access it at all. Establish a governing committee made up of functions from across the organization, and develop a "common business language", an agreed upon definition of key data points and important business metrics. This will achieve that single source of the truth, and ensure that decision makers spend time making decisions rather than arguing over definitions.

Formalize the process of answering these questions, and document results. This way, several years from now, when everyone who originally asked for and evaluated the data have gone, you know why it's there, who is using it and what the impact of changing it would be.

2015 will be the year of (new) technology Data and technology have always gone hand in hand. If you want the data, you need to have the technology platforms to collect, store, integrate and process that data. Evolutions in storage and processing power have probably had the biggest impact on revenue management analytics – and profitable pricing strategies – in the last few years. It is crucial to have a working knowledge of the technology enablers that underpin innovations in systems and analytics in hospitality. I am not suggesting that revenue managers need to become IT experts as well, but it is important to have enough knowledge that you can intelligently participate in the discussions with IT counterparts, and understand the implications on your business. We all know that bad technology decisions can be among the most expensive failures in hospitality, so now is the time to "reach across the aisle" to partner with IT counterparts.

Here is What You Need to Know About Technology in 2015

Data storage – The reason why big data has become such a big deal is not just because there is so much of it, but also because it is in different formats than we've traditionally had to deal with. Sales data or customer profile data is structured data. There are clear relationships among data sets and sources, and it can be neatly sorted and stored in rows and columns. Experts suggest that 70-90% of all data created today is unstructured data, like text, click stream, video, or audio. This data cannot be neatly sorted into rows and columns, but it does contain valuable insight about the business. Databases have had to evolve to handle this unstructured data. This is why you will hear many technologists talking about Hadoop, which is a database specifically designed to facilitate storage of and access to unstructured data. Hadoop, or similar technologies, become valuable when you want to take in a lot of new sources of unstructured text data for analysis, but not be required to enforce a pre-determined structure.

Most of the data required for revenue management analytics is structured, and the forecasting and optimization analytics in revenue management systems do required a pre-determined structure – even if the data started as unstructured data. Hadoop, or any unstructured data platform, would likely not be the primary data source for a revenue management system, but revenue managers may still want access to unstructured data to do ad-hoc, strategic analysis (like unstructured text data from reviews for example).
Data processing – Processing power is all about speed to answers. Since forecasting and optimization algorithms require many iterations of complex mathematics, they are processing-intensive. In the past, compromises were made (summary data, heuristics instead of true optimization) to ensure that the algorithms could run overnight and prices could be changed at least once per day. Now these compromises are no longer necessary. Processing power has increased to the point that optimizations that used to take days are now solved in minutes. This has been accomplished through several innovations in analytics execution.

At a High Level, There Are Three Methodologies That Increase Speed to Answers

  • Grid computing, or parallel processing, splits problems into sections and solves them simultaneously, then consolidates the answers (picture the speed difference between you adding 8 numbers or having four friends add two numbers each).
  • In-database processing saves time by reducing data movement. The analytics go to the data, instead of moving data into the analytics application and back again with the answers.
  • In-memory processing saves time by reducing movement on a physical disk. It is slower to read data off of a hard drive or disk, so the processing happens in the computer's memory. As memory becomes larger and less expensive, in-memory processing becomes more powerful and faster.

Increases in processing power will facilitate dramatic improvements in revenue management analytics, including the ability to apply multiple forecasting methods to increase forecast accuracy, and "re-optimization on the fly", to support what-if analysis.

Data visualization – The same advances in processing power that help facilitate faster analytics also support faster reporting. Billions of rows of data can render in fractions of a second, even with drill downs. Increased flexibility in the tools facilitates self-service reporting. Instead of having to wait for IT to create a report, business users can create their own visualizations with drill downs and explorations. All of this puts the power into the hands of the business users to access the data they need to make decisions at the speed of business.

There is plenty of information out on the web about all of these data and technology innovations. My goal was to give you some exposure to what you will start hearing, as well as a sense of the possible implications on your job. The better informed you are about opportunities, the more prepared you will be to work with counterparts to prepare your organization to survive and thrive in 2015 and beyond.

Reprinted from the Hotel Business Review with permission from www.HotelExecutive.com

Angela Lipscomb
919-531-2525
SAS Institute Inc.