Most organizations conduct guest satisfaction surveys to determine the guest’s levels of satisfaction, loyalty, or perceptions of employees or facilities. However, we often don’t know how these measures relate to financial metrics such as revenue, and base many operational and revenue choices and decisions on assumptions made about those relationships. For example, we often make the assumption that guests who receive comp rooms create more revenue on the gaming floor than guests who do not receive comp rooms. But do we know that? Can we show evidence that the comp dollar returns to the organization through a different channel? By integrating property-wide data sets, which are often disparate and siloed, you can determine the answer to this and many other questions.
Not only has there been lack of transparency and sharing in many internal organizational data sources, the link between guest survey data, other guest data and financial metrics has been long-neglected, but it is remedied simply. This article will discuss steps you can take to integrate your property-wide data sources and present some options for its use.
There are a few general steps to take to begin the integration process. The first and most important step is to create a unique identifier that is attached to each guest ate very possible data point including spending in various centers within the property, and the guest survey. This identifier must be unique from the guest’s loyalty card number, in that the identifier will represent both the individual guest as well as that unique visit time and date. Some properties may use the loyalty card number plus a date code or other specification to compose the identifier. Each occurrence of data within the entire system must use the same identifier for every transaction or guest survey by the same individual. It is uncommon for an organization to have databases shared across the entire organization, but having each functional area use the same identifier will allow all data to be integrated on a one-to-one basis. The more data that can be linked together, the more complete the picture of your guest and their behavior will be.
Once the unique identifier is inserted into all data sources, analysts can integrate all data using commonly-available programs such as SAS, Microsoft Access, or SPSS from IBM. The data integration involves creating data relationships between all databases based on the unique identifier, then integrating all available data, typically as additional variables. This integrated data set allows you to accomplish several objectives including customer segmentation, determining differences between groups and creating predictive models for revenue in various areas of the property.
Integrating all data sets allows you to segment customers based on many factors above and beyond the standard demographics, spending level, or visit frequency. Integrating guest survey data with financials and spending in all areas of the property allows segmentation based on attitudes (from the guest survey) and behaviors (spending behavior in all areas). Demographics are frequently inaccurate for purposes of segmentation and prediction. More specific attitudinal or behavioral customer segmentation provides improved customer targeting, thereby improving the efficacy of marketing and guest relationship management.
This comprehensive data integration also allows you to separate guests into logical groups to determine if there are important differences between groups. For example, do leisure travelers spend more in retail outlets than convention goers? When customers check in at the front desk, is there a measurable increase in upsell revenue? Do guests who receive comp rooms spend more in other areas of the property? Separating customers into those who rated front desk satisfaction a 9 or 10 (the highest rating), are highly satisfied customers staying on property for more activities such as dining, shows and clubs? The answers to these types of questions can inform your marketing activities (especially for targeting promotions), operational improvements and revenue management.
Finally, the integrated data set can be used to develop a predictive model. Predictive models can show which factors are leading to the result of interest, such as cash revenue, gaming spend, retail spend or return visits. With the integrated data set, these predictors can include perceptual variables such as guest satisfaction, intent to return, or feelings and emotions surrounding their visit. These predictive models are easy to set up in programs most analysts are familiar with and mean much more than simple correlations. The predictive model can predict, with an identifiable level of error, to what extent a specific level of predictor leads to the outcome of interest. This information is also essential for determining the importance of operational performance, revenue management decisions and marketing expenditures in terms of accomplishing goals of interest, such as revenue levels.
Many decisions made about guests and customers are based on assumptions. Those assumptions are often based on extremely valuable experience and knowledge. Data can be used to support that qualitative source of wisdom and identify areas in which guest or customer preferences may be more dynamic. It is not common for data to be shared across functional areas particularly in integrated resorts. However, this is an easy problem to remedy by incorporating a unique identifier into each functional area’s data collection. This unique identifier allows you to integrate guest satisfaction, spending, and revenue into a single data set, thereby creating an opportunity to create the full picture of each guest on an individual basis. This individual description allows more customized marketing to be developed, leading to increased conversion on marketing spend and can guide other operational and revenue management decisions.
Dr. Heather Monteiro is the Market Research Analyst for the UNLV Hotel College’s Center for Professional and Leadership Studies (PLuS Center). Dr. Monteiro earned an MBA from Georgia College and State University and a Ph.D. at Georgia Southern University in Logistics and Supply Chain Management, with a minor in Marketing/ Consumer Behavior.