Pricing

Carlson rerecord hotel group maximizes revenue through Improved demand management and price optimization Presented To: Professor Verve Talking Table of Contents Introduction Revenue management is the art and science of selling the right product to the right customer at the right time for the right price, through a combination of Inventory controls and pricing (Cross 1997). Traditionally, the Carlson Render Hotel Group (CRAG), the world ninth-largest hotel group by revenue, utilized a revenue management system (ARMS) that controlled availabilities for deferent categories of products they offer, typically using Length of Stay (LOS) control.

For examples, for a specific room type, one future arrival date might have an associated full pattern LOS= 1 Nights. Thus the decision was, whether each LOS should be open or closed for sale on that arrival date. This model was difficult for time-pressed hotel operators to understand. Moreover, it has to be manually done by the operator and thus its very time consuming. As a result an arising need for a more sophisticated and more efficient model was requested by CRAG.

The changes In operating conditions that happened in 2006 for the hotel industry, had shifted the ARMS from LOS control to rice control which was considered as the prominent differentiator between hotel rooms, Ignoring elements of the room offerings. The challenges in this new system, were to build a model that accounts for the price sensitive nature of demand, and put the competitor data at the heart of its forecasting and price recommendation process.

Moreover, the implication of this solution must address the large and complex hotels, and smaller hotels with limited staff, and during all types of stay nights. Solution and Methodology The CRAG collaborated with JDK software group on a highly innovative revenue optimization project, Stay Night Automated Pricing (SNAP). The project’s methodology can be splinted into 3 major parts: Enterprise Demand Forecasting They defined the lowest level of forecasting, demand forecasting unit (UDF), by hotel, rate segment, LOS, day of the week, and booking interval.

The utilized Multiple Linear Regression techniques as the forecasting technique, where demand is expressed as a linear function of a level term representing De-seasonality demand, Fourier Series seasonality term, and binary causal factors for holidays and special events. In this del, a hierarchical forecasting approach was followed by defining high level (parent) DIF and low level (children) DIF which are interrelated. 1 OFF This part is based on price elasticity of demand, competitor rates, remaining room inventory, demand forecasts, and business rules.

A continuous quadratic programming model is utilized, where the stay night prices are optimized subject to arrival dates constraints. It assumes LOS to be a sum of the prices of each consecutive stay night spanned by the arrival date and LOS. A network optimization formulation is followed because each stay night and LOS and the demand for a given eight is the sum of the demand from arrivals on that date and arrivals from previous dates that has not yet checked out. Finally, the solution must keep the rates in line with the marketplace.

Technical Development and Implementation IDA created a productized solution for implementing the prototype called, Travel Price Optimization (TOP). TOP facilitates the user workflow and provides the necessary tools to help the user to understand and evaluate price recommendations through: 1 . Demand View: Compare expected number of rooms predicted to be sold at current and recommended rates 2. Rate View: Compare current and recommended rates against competitors’ rates 3. Inventory View: Provides an idea about allocation of room inventory of different types 4.

Rate mix View: Highlight the changes in expected booking to come in each rate segment 5. Rate Curve View: View the trajectory of booking at the current versus recommended rates from the current days left through to the stay night Conclusion and Future Plans The SNAP project had allowed CRAG to efficiently and effectively conceive a vision for demand forecasting and price positioning with respect to competitors, which will enable it to better project bookings, cancellations and associated revenues.

In such cases, a spill-based recommendation would simply recommend pricing Just above marginal cost. Moreover, its revolutionary aspect had allowed hoteliers to objectively measure price elasticity of demand, incorporate competitive rate data to be used as the basis to set price, and capture the customers’ willingness to pay. Implementing this project, had increased the revenue by 4-5 % in the year of 2010, this value varied between one hotel and the other based on the degree of compliance of the hotel manager with the recommended prices.

Finally, this solution had made pricing consist across all hotels of CRAG, and increased the transparency of distribution channels which allowed the customers to price shop across all channels. However, price recommendations did and still do face some resistance from hotel managers, especially when it significantly deviates from the historical and competitor prices. As a future plan, CRAG wants to develop a group evaluation module (GEM) as an extension to the TOP. GEM recommends a walkway rate that is the minimal acceptable rate for a group request based on the amount of transient demand that would be displaced by that group.