You Might Want to Sit Down for This: Modeling Consumer Seating Preferences

Journal of Marketing Research Scholarly Insights are produced in partnership with the AMA Doctoral Students SIG – a shared interest network for Marketing PhD students across the world.

Picking seats for a movie can be a tough decision. Some people prefer the proximity of others, while some like to maintain a bit of distance. Others prefer aisle seats for a quick exit, while some target more central locations for the optimal view. Given the variety of motivators for picking a specific seat, firms face a daunting challenge when trying to optimizing their business practices to account for these heterogeneous seating preferences.

To maximize performance outcomes such as occupancy rates, revenues, and profits, firms with seating venues must take strategic actions to account for consumers’ varying preferences. To this end, firms have sought to utilize advanced ticket seating maps and variable pricing based on consumer data to improve their performance. In a recent Journal of Marketing Research article, Simon J. Blanchard, Tatiana L. Dyachenko, and Keri L. Kettle develop a novel measurement approach that guides operators and managers to make more effective seat-design decisions based on a variety of consumer preference variables. Their unique model focuses on and emphasizes the need for individual- rather than aggregate-level seat preference data, to better account for consumers’ preferences. The authors also illustrate the use of their approach in four experiments to show how their model can be leveraged to optimize performance outcomes for seating venues.

Inspiration from a hockey game

We asked the authors how the idea of their paper originated and learned that it was from their personal experiences. After attending a world cup hockey championship game in 1996, one of the authors noticed that some of the seats in the first row of the arena were priced the same as seats several rows back. This experience piqued his curiosity and fascination for quantifying seat preferences in various settings, and it spread to the rest of the author team. During the research process, they quickly learned that seating venues typically account for key locational characteristics in ticket pricing strategies (e.g., the view, aisle seats, type of event); however, individual-level seating preference variables were often overlooked. From their research, the authors discovered that individual-level preferences such as the distance from key focal elements (e.g., screen, theater platform) and other filled seats varied significantly across consumers. Therefore, they developed a model to account for these individual-level preferences to optimize seat prices and availability, helping firms maximize consumer satisfaction, revenue, and occupancy. Also, operators can use their model to run simulations of seating choices (see, for example, seatmaplap.com) to test new seating strategies.

The model, its applications, and generalizability

In the authors’ proposed model, it is suggested that venues can withhold or temporarily show seats as unavailable to maximize occupancy. This tactic can be useful for venues looking to optimize their seating performance, along with similar tactics such as releasing tickets for specific seats at different points in time and dynamically pricing seats. Currently, seating venues withhold seats at an aggregative level; the paper’s model expands on this practice by providing an avenue to assess the effectiveness of withholding seats at an individual level. Seating venues can then show seats as unavailable on an individual basis to optimize seating performance outcomes. This may cause some concern if some consumers see this tactic as unfair; therefore, the authors cautioned that personalizing unavailable seats must be done carefully and in a nondiscriminatory manner. However, they noted that these policies’ long-term effects would be an excellent avenue for future research.

Venue Shape. The proposed model features the measurement of locational preferences within a rectangular-shaped venue. However, we wondered whether the model could be easily applied to venues featuring other shapes. The authors told us that their model could be easily adapted to accommodate differently shaped venues and venues with different types of focal elements and varying sizes. Therefore, the proposed model applies to movie theaters and sports arenas, outdoor concerts, and many other differently shaped seating venues. Suppose one wanted to implement the model for a large sports stadium. In that case, the model could be adapted to account for the consumer’s two-stage decision process: (1) the choice of section and (2) the choice of seats within the section. Other focal points can also be taken into account based on angles and distances from various focal elements. This flexibility allows the model to be utilized in various situations while the firm’s decision makers identify key focal elements and characteristics.

Varying Demand. It is also important to point out that the individual-level choice model was not originally developed to account for event-level demand. In other words, the model does not take into consideration the final occupancy, which is a variable used in box-office prediction models. The model instead focuses on why a consumer selects specific seats at an individual level (if any). However, the authors conducted further experiments to explore the model’s generalizability concerning forward-looking consumers. They would expect seats to be filled even if they are available at the present moment. The experiments were run with varying amounts of time before an event and with varying levels of current occupancy proxying the expected popularity. The results showed that when 80% of seats were reserved three months in advance, consumers would select seats based on their location to environmental elements, such as the screen and aisles instead of personal space. This makes logical sense and further expands the generalizability of the model concerning event-level demand.

Alternative Focal Points. The model’s generalizability also made us curious about venues with no focal area of attention, such as public transportation and restaurants. When we asked the authors, they reasoned that every environment has structural elements that provide or detract utility from the consumer’s proximity to them. They argue that in such instances, the focal point changes depending on consumer interests and preferences. For instance, in a restaurant, consumers might gravitate toward better views and away from overly busy areas like entrances and doors. On airplanes, consumers might prioritize exit seats for security purposes or back seats to access the restrooms on small to medium planes. The authors note that the key commonality in these scenarios is that consumers will have varying desires of proximity to focal areas and that operators should systematically investigate their assumptions about these consumer preferences.

Key Implications

To sum up, this study makes unique research contributions revolving around advanced selling, spatial models, and personal space, focusing on an individual-level measurement approach to determine how consumers favor locating themselves regarding focal elements. It also sheds light on the heterogeneity of consumer choices to be located near focal event elements and the other consumers. For event operators, this means expanding data collection beyond ticket sales by including the consumers who did not purchase tickets. For managers, this study provides an optimized model to aid in the use of fitted individual-level parameters in more effective decision-making regarding seat-level availability.

The Impact of COVID-19

The authors further elaborated on the implications of the COVID-19 pandemic for their research. In particular, we asked them about the specific adjustments that would need to be considered when applying their approach to venues enforcing social distancing. The authors addressed this question from a methodological and strategic perspective, arguing that their study provides an excellent first step in systematically helping operators learn more about consumer preferences for personal space during these times.

From a methodological perspective, the authors state that few adjustments are required to implement their model. However, they highlight the importance for operators to find ways to improve consumers’ confidence with respect to having sufficient personal space during social distancing enforcement. This can be achieved by adaptive seating charts that instill this confidence when presented to consumers.

From a strategic perspective, since regulation-compliant seating-position data is not yet available, the authors posit that their experimental research approach could provide operators with insights on seating arrangement options. Questions like “Which assortment would lead to the fewest decisions to skip the event?” or “Which assortments have locations that are likely to be infrequently chosen?” could be answered using this approach. Thus, the lack of field data could be mitigated using this approach’s experimental nature and can be used to build confidence in predicting consumer preferences.

Article Referenced:

Blanchard, Simon, Tatiana L. Dyachenko, and Keri L. Kettle (2020), “Locational Choices: Modeling Consumer Preferences for Proximity to Others in Reserved Seating Venues,” Journal of Marketing Research, 57 (5), 878–99.