Exploring the potential of dynamic pricing for fixed-capacity events

Using regression modeling on Belgian professional football league sales data

What is the potential of dynamic pricing in optimizing ticket sales of fixed-capacity events such as sports events and pop/rock concerts?, That was the question facing Master’s Degree student Gilles Gerlo of Ghent University. Using SAS software, he investigated the case of Belgian football’s professional league. He took inspiration from dynamic pricing strategies introduced to the Major League Baseball season in the United States a few years ago. These strategies had proven to be a great success. Gerlo’s regression modeling analysis identified some potential but also indicated that real-life test programs should be run to predict which strategies would work best for a given football club.

It will be necessary to run a comprehensive test program over
an entire season to determine which price dynamics strategies
will be most effective.

Gilles Gerlo
Master’s Degree candidate Business Analytics at Ghent University

An opportunity for clubs and fans

Today, dynamic pricing is common practice in the aviation and international railway sector. Here ticket prices are constantly changing based on continuously evolving information such as seat availability, actual demand and time of sales. However other sectors affecting a fixed-capacity offering seem to be lagging behind. A prime example is the Belgian professional football competition, where ticket prices are fixed at the start of the competition season. “It seemed to me that clubs were missing an opportunity for both fans and owners,” says Gilles Gerlo. “Dynamic pricing may have the potential to boost sales and therefore the fan base as well as increase club revenue. Fans profit as well. Putting the right strategy in place might for instance lead to cheaper tickets when purchased months in advance and gradually raising prices until capacity is reached. Other strategies are also possible which take into account other key dynamics. These can include such factors as the constantly changing competition ranking of teams as the season progresses and even weather forecasts.”

Learning from San Francisco

Thesis advisor Professor Dr. Dries Benoit suggested taking inspiration from recent advances made by Major League Basebal in the United States. In 2009, the San Francisco Giants (SFG) started a dynamic ticket pricing program in close collaboration with ticketing software firm Qcue. “Dynamic pricing proved to be very effective there,” confirms Gerlo. “They established a regression model to change ticket prices based on a massive amount of variables, including competition particulars, featured players, and even time of day. Most importantly, they had data from sales on secondary markets such as eBay, giving an indication of the price fans were willing to pay for a given seat at a given moment. Initially SFG management feared negative reactions by fans to dynamic pricing. Therefore, benefits such as substantial discounts for early buyers were essential in the PR campaign implemented for dynamic pricing. The end result was very positive. In recent years, SFG have almost always sold out completely and revenues have increased by 2.8%.”

Regression model established

When examining the case of Belgian professional football during the 2012-2013 and 2013-2014 seasons, Gerlo immediately understood that there were important differences. “The number of games is significantly smaller,” he says. “This means that we have less available data to construct a reliable model. In addition, available data indicated that ticket demand is rather inelastic in many of the football clubs. Some clubs already have a near full house for every home game. In this context, the potential for additional sales through price reduction is very limited.” Nevertheless, Gerlo was able to construct a ticket sales regression model with potential predictive power. He based the model on ten clusters of variables, including circumstantial indicators such as demographic data and occupation rates in previous games as well as performancerelated indicators such as number of goals scored and current rankings in the race for the championship.

Potential needs to be tested

The explanatory strength of the ticket sales model greatly varied. “Explanatory strength was as low as 24% for Club Brugge, where fan loyalty is very high,” says Gerlo. “On the other hand, the model scored as high as 93% in explanatory strength at AA Gent, where occupation rates went from low to high and back again during the seasons beings analyzed. It is clear that for clubs like AA Gent the model has potential in predicting ticket purchasing behavior and therefore in dynamically adjusting prices to improve sales and occupation rates. However, it will be necessary to run a comprehensive test program over an entire season to determine which price dynamics strategies will be most effective. I encourage the clubs to pick up the gauntlet.”

Universiteit Gent


Investigating the potential of dynamic pricing to optimize ticket sales in Belgian  professional football league


SAS® Enterprise Guide

Enjoying the power of SAS

Gilles Gerlo was very pleased to use SAS software for his analyses: “The software provides tools for all of the advanced analytics methods that I needed, including variable clustering and regression. In accordance with the policies in place at my faculty, I used SAS code exclusively, rather than the available business tools. This is fine of course, in an educational context where one has to learn the ins and outs of analytics. Luckily, I was also awarded a one month traineeship at SAS Institute, which enabled me to discover the power of tools such as Enterprise Guide.”

The results illustrated in this article are specific to the particular situations, business models, data input, and computing environments described herein. Each SAS customer’s experience is unique based on business and technical variables and all statements must be considered non-typical. Actual savings, results, and performance characteristics will vary depending on individual customer configurations and conditions. SAS does not guarantee or represent that every customer will achieve similar results. The only warranties for SAS products and services are those that are set forth in the express warranty statements in the written agreement for such products and services. Nothing herein should be construed as constituting an additional warranty. Customers have shared their successes with SAS as part of an agreed-upon contractual exchange or project success summarization following a successful implementation of SAS software. Brand and product names are trademarks of their respective companies.

Back to Top