Skip to main content

Demand Forecasting Success

Hotel companies have used revenue management (RM) technology for decades to drive incremental revenues. Most operators have achieved impressive financial results through RM, but many have failed to capitalize on their market opportunity. At Carlson Hotels, we took a hard look at the factors influencing successes and failures in RM, and we arrived at the conclusion that forecasting is by far the most important -- and challenging --element of good revenue management. Let's be clear: by "forecasting" we are not referring to a report; we are talking about a process that enables hotels to manage inventory, rates and -- most importantly -- expectations about the future. You cannot alter the future of your business without strong estimates of what that future actually holds. This means hoteliers that base their operations on a strong demand outlook are the most successful with RM.

Our goal at Carlson is to deliver technology- enabled revenue optimization capabilities that will make our hoteliers more successful. Since we know that understanding and predicting demand is the hardest piece of the puzzle, we decided to build a centralized capability so that our hoteliers don't have to.

The right tools
In March 2007, Carlson launched its revenue optimization project, starting with an enterprise-wide implementation of JDA Software's Demand solution (, which had never before been implemented in a hotel group. Our analysis of existing hotel RM systems showed that while most have good forecasting capabilities, their functional limits are strictly defined.

Modeling capabilities were particularly important to us because we are predicting demand for hotels that may be located thousands of miles away from our forecasting experts. That presents the challenge of integrating local market factors into our centralized demand forecasting process.

We examined how hotels manage this process and found that the most successful revenue managers build a yearly event calendar using local market information and estimate the impact that each event will have on demand. Some automated RM systems capture historical booking statistics relating to past events. For example, they may use the booking history from Memorial Day 2007 to predict the demand pattern for Memorial Day 2008, but the process takes time and hard-pressed revenue managers cannot always execute it thoroughly.

This limitation has seriously compromised the RM initiatives of many chains and few have found an effective process to capture local market intelligence. Carlson's search for a more effective method that would lighten the load on our hoteliers led us to an approach that is new to the industry. We engaged a third party, MKT Data Services (www.mktdataser, to provide event data on all the markets that we serve.

Powerful knowledge
MKT's database contains millions of events ranging from major league sporting events to book festivals to concerts. It offers the ability to categorize, so we can identify specific events, even when they occur on dates that differ from historicals. For example, imagine you're running a hotel in Madison, Wisconsin where the college football schedule is one of your biggest demand drivers. This year, four of the first five games will be at home, but that is the inverse of the previous season. This changes the demand outlook for the city. Now, we can pinpoint these and many other demand-influencing factors in each market where we have hotels, and we can do it automatically. But to benefit Carlson's hoteliers, we had to predict the impact on their demand in each market from every event -- and this is why we turned to JDA Demand for advanced modeling capabilities. Today, Carlson is incorporating MKT's regional event information into its demand forecast in a fully automated environment, and we are seeing an impressive improvement in the accuracy of our demand outlook based on the addition of regional event data.

You can never replace local knowledge in any business, but by using information available in the public domain and some cutting-edge forecasting capabilities, our centralized team can now deliver insights to our hoteliers that help them make better decisions and ultimately more money. Revenue optimization is a core competency for any good hotel chain, and with a little help from some advanced technologies, it can more than ever contribute to a hotel's success.HT

More Blog Posts in This Series

This ad will auto-close in 10 seconds