Historical demand has long been a critical component of dynamic pricing and Revenue Management Systems (RMS). An even more critical requirement for RMS is the unconstrained demand, which is the true demand for a particular product in the absence of any limitations, such as when a room or seat is unavailable to purchase. The success of unconstraining affects the entire pricing and revenue management process.
Attempts to Use Regrets and Denials in Unconstraining Have Been Largely Unsuccessful
There are two options for unconstraining the demand: direct observation and recording or the use of statistical methods. Since the adoption of RMS in the hospitality industries, the unconstraining methodologies have been primarily analytically-driven and attempts to use real-world or directly-observed customer data, specifically regrets and denials have been largely unsuccessful.
Regrets and denials data are not directly applicable to calculating an unconstrained demand since there is an important distinction between “denials” that are due to unavailability and “regrets” that are due to price or other factors. Many reservation or booking systems are unable to automatically capture the difference between regrets and denials.
The Unqualified Transient Data from Brand.com Is Insufficient
There have been recent claims of RMS systems incorporating denials data in their algorithm using brand.com data, which merely captures partial regrets and denials data for the unqualified transient demand. However, predictive models using only the unqualified transient disregard the demand for different market segments and/or additional channel behaviors. The problem with this methodology is that not only brand.com comprises only 27% of the reservations for transient nights, as TravelClick reports, but also unqualified transient demand data is being widely used without sufficient regard for unconstraining. Unconstraining methods must include demand for each and all of wholesale, group, corporate negotiated, and unqualified transient demands. This is what we call ‘holistic unconstraining.
A Guest Searches Multiple Times for the Same Room
Property managers or reservations personnel may also be unaware of the guest’s booking history. For each call, the denied availability will be coded independently when, perhaps, they are related to a long list of inquiries from the same person.
Regrets and Denials Data Is “Dirty Data”
Various news media have reported that, while bookings have remained generally flat, booking engine transaction volume has increased substantially. This is a strong quantitative confirmation of what is known anecdotally: the look-to-book ratio is extremely high, and continuing to increase. Studies confirm that many travel buyers use a variety of websites to research and compare prices before making booking decisions. So, even if one is confident about the methodology for assessing denials on one site, it is not at all clear that there is not cross-usage of additional websites or multiple visits per buyer that are unknown in the denial logic. That is primarily why leading data scientists refer to regrets and denials as “dirty data.”
Each of the scenarios above can result in the RMS over-unconstraining the demand data, which leads to over-protection of inventory and, eventually, reduced occupancy.