November 2008

Data mining within the hotel industry
The role of data mining and analytics can be significant in helping hotel operators understand their customers’ diverse needs. By Richard Boire

Our last column focused on how the sports industry could use data mining techniques as a competitive business advantage. We also examined how marketers can identify unique groups (such as sports fans) and use intelligence about them to develop appropriate strategies and programs. This also works for other types of customers. In this article, we will explore how data mining can be used in the hotel industry.

Historically, hotel customers’ needs were quite narrow, with mobility being the key limiting factor. Back then, the population was arguably more homogenous and had similar needs. Things have certainly changed. In today’s complex world, people of all different cultures and tastes travel widely and have diverse needs and interests wherever they stay.

What they all have in common, however, is the view that a hotel stay is “an experience” rather just a visit. Such activities as fine dining, nightly entertainment, spa sessions, and corporate seminars and meetings nurture this notion. It is easy to understand that such a range of activities will have varying levels of appeal across a given clientele. The role of data mining and analytics can be quite significant in helping hotel operators to better understand these varying client needs.

Seasonality matters
The first task might be to conduct a basic customer value exercise in order to ultimately identify our best customers. But, as with many analytical exercises, the concept of seasonality needs to be considered here (which arguably is even more significant within the hotel industry). For many projects, we make the assumption that the rank ordering of customers is unaffected by seasonality. In other words, customers may spend more during different times of the year but their spend relative to each other remains unchanged. Depending on the industry, some analysts could certainly debate this notion but most would agree that for the travel industry in particular, the issue of seasonality has the potential to significantly impact travel behaviour. For example, I may spend $1,000 annually as a casual traveler throughout the year and am considered an average customer. Meanwhile, my sister spends $1,000 annually but on a tennis package for one week in August. Should she be considered an average customer? Hardly. She is a better than average customer if we factor in the seasonality component of her behaviour. We both spend the same amount but are, in fact, very different types of customers. This notion of seasonality is significant when conducting any analytics exercise, particularly, if we consider that many hotels will offer tennis and golf packages in the summer and ski packages in the winter.

Besides the issue of seasonality, there are various services that may have more appeal to certain groups of customers. Fine dining and theatre may appeal to one client, while spas and valet type services appeal to another. With varying interests amongst travel clientele, a cluster type segmentation exercise would be a very useful. Experts with domain knowledge in the travel industry would certainly understand that there are distinct or homogenous customer segments. By applying the data captured from travel customers, we can now apply some science to identify what are the truly distinct customer segments.

So how do we integrate the notion of customer value within the cluster segmentation approach? In many typical data mining exercises, we might conduct a value segmentation exercise on the entire customer base and then overlay the cluster segments to see how they align with value, as illustrated in the following chart. (See Figure 1.)

Figure 1. A value segmentation exercise
In this example, the data tells us that high value customers comprise mainly Cluster 1 type customers while low value customers primarily comprise Cluster 3.

Figure 1. A value segmentation exercise

But we might ask ourselves whether this approach is appropriate for the hotel industry. A more useful approach might be to identify the unique cluster groups and to then conduct a separate value segmentation exercise for each cluster. Let’s say that for a given hotel, we arrive at four simple key cluster or distinct customer groups such as tennis group, ski group, pampered group (use spa and valet type services) and the nighthawk group of fine dining and theatre goers. The segmentation approach might look as follows (See Figure 2) for this group.

Figure 2., A value segmentation exercise for each cluster
Here, we can see that the tennis group has the lowest value of the four customer segments. In fact, the highest tennis value segment might be considered of average value for a pampered group or nighthawk group customer. This tiered segmentation approach for each cluster allows us to evaluate each group separately on its own merits.

Figure 2. A value segmentation exercise for each cluster

Applications for this information
As seen in the previous charts, initial learning from this type of segmentation could be used in developing a marketing strategy which is data-driven. For instance, regarding the ski group cluster, some top line thoughts on how we might use this information include:

• Cross-selling tennis packages to all ski package customers while cross-selling ski packages to tennis package customers
• Up-selling getaway type packages which include fine dining and a theatre pass to high value ski and high value tennis customers
• Cross-selling a spa type package to high value nighthawk customers (deciles 1,2, and 3)
• Providing President’s type services and benefits that are exclusively available to a very small elite group of customers—that are both high value pampered (deciles 1,2, and 3) and high value nighthawk (deciles 1,2, and 3)

Commercial trade
Besides consumer activities, we cannot forget the commercial sector and its needs within the hotel industry. The hotel could examine its current commercial customer base and once again establish unique groups of customers. For example, we know that there are groups of commercial customers which simply use the hotel for overnight stays while other groups are there for seminars or special events. In fact, these above groups could be further segmented based on industry sector. We would certainly expect that seminars geared to oil and gas might be more appropriate in Alberta with financial services type seminars being more prevalent in Toronto. Of course, all this supposition on what might define the unique business segments would again be determined quantitatively through clustering routines. By using the data and mathematics to define these segments rather than intuitive judgment, we can once again develop unique programs that are geared to different groups of unique business travelers.

Customer experience has always been the overriding customer philosophy within the travel industry, long before the advent of data analytics. Yet, with data analytics, the travel industry can now use information to make better decisions regarding its customers. This enhanced decision-making capability from the data enables the specific hotel to be more proactive with its customers. Traditionally, success in the hotel industry has always been determined by superior customer service actions that address the immediate requested needs of the customer. The competitive advantage in today’s world will be the ability to meet the unrequested needs of customers which can only be determined through data analytics.

Richard Boire heads up the Boire Filler Group, a Pickering Ontario based firm specializing in customer database analytics and predictive modeling. He can be reached at 905-837-0005.