Hospitality executives constantly struggle to achieve balance between maintaining a memorable guest experience and meeting revenue and profit obligations to shareholders and stakeholders. Err too much on the side of guest experience by offering too many complimentary amenities, for example, and profits can suffer. But, cutting labour to save costs can create long waits at check in, and the guest experience is damaged. Analytics can help strike a balance between these two critical aspects of hospitality operations.
Analytics are broadly categorized into two main groups: descriptive and predictive. Descriptive analytics are also called “business intelligence.” These analytics describe what is happening in the data through metrics such as percentages or averages, typically displayed in static reports or dashboards. Descriptive analytics answer questions such as “how many, where?” or “what happened?”
Predictive analytics anticipate trends and foresee opportunities. They use historical data to predict the future, answering questions such as “what if these trends continue?” or “why is this happening?” and “what’s the best that can happen?” These analytics allow managers to proactively plan for the future, hedging against risk and taking advantage of opportunities. Predictive analytics move the organization from reactive decision-making to proactive decision-making.
There are several categories of predictive analytics:
• Statistical modelling: understands why trends in the data are happening by identifying which factors have a relationship with each other and how much they influence key measures
• Predictive modelling: uses historical data and key demographic variables to predict behaviour or outcomes. used when there is knowledge of the relationships or the format of outcomes
• Data mining: detects patterns in large data sets and either describes the patterns or uses them to predict future outcomes. Used when operators don’t know what kind of data they’re looking for, or what outcome they are expecting
• Forecasting: analyzes historical patterns and current conditions to predict a future state (revenue, demand or guest counts, for example)
• Optimization: calculates the best possible answer or outcome considering operating conditions and business constraints
These predictive analytic techniques are either delivered in an analytic tool that allows for ad-hoc analyses, or built into a solution with pre-defined inputs, methods and outputs. Most organizations utilize both delivery methods, depending on the problem.
Analytics can be used in key functions of the hotels:
Pricing and Revenue Management
Revenue management is generally the most data- and analytics-intensive department within the typical hotel. To accurately price rooms, revenue management systems forecast demand, and then optimizes the price and availability of rooms to maximize the revenue from the limited capacity of hotel rooms. Within this process, statistical analysis is used to model no-shows and cancellations, “unconstrain” demand and calculate price sensitivity. At this point, most hotels use a revenue management system that has been specifically designed to execute the complex analytics required to deliver an optimal price. The analytics processes are configured to the hotels’ specific operating conditions and connected to selling systems to deliver price and availability controls.
Marketing and Customer Loyalty
Marketing and customer loyalty are quickly following in the footsteps of revenue management when it comes to utilizing predictive analytics. To develop a better relationship with the guest, segmentation and profiling models are used to group guests in segments that have similar characteristics, whether they are business-defined segments or demographic or behavioural defined segments. Return trip models are used to calculate the probability of a guest returning to the property in a specific period of time. Lastly, one of the most widely used predictive techniques for marketing and loyalty calculates the customer lifetime value, determining how much a guest is worth during the expected lifetime of his/her relationship with the company. Understanding a guest’s predicted lifetime value can help determine how to treat guests, including offering incentives in the form of promotions and discounts.
Operations
Forecasting tools are particularly valuable to operations. Accurate demand forecasts can support labour scheduling and supply ordering. Revenue forecasts assist budgeting and planning. Statistical modelling can be used to understand the drivers of guest satisfaction. Text analytics interpret the content and sentiment of reviews and open-ended guest survey questions to identify service improvement opportunities or design new offerings. Optimization can produce a labour schedule that minimizes labour costs while maintaining service levels. There are some automated systems, but many hotel companies also utilize tools for ad hoc analysis.
Today, many hotels are broadly implementing visualization tools for descriptive analytics. Revenue management departments are heavy users of advanced analytics in their systems, and some marketers are applying basic segmentation analysis or customer value calculations. Those hotel companies that are striving for a competitive advantage will go beyond descriptive tools to apply advanced, predictive analytics, moving the entire organization from reactive to proactive decision-making.
By Kelly McGuire and Natalie Osborn, SAS Institute
Kelly McGuire and Natalie Osborn are hospitality industry specialists at SAS, the world’s leader in business intelligence and analytical software. To learn more visit the analytics hospitality executive blog where McGuire and Osborn write about solutions to the hospitality industry challenges.
