The Spokane Symphony has developed a new model to guide booking decisions and forecast ticket sales at The Martin Woldson Theater at the Fox, in downtown Spokane.

Jeff vom Saal, executive director of Spokane Symphony, says the system, known technically as a multivariable regression approach, has a great deal of potential to help the Fox find financial stability by selecting performers that, based on mathematical forecasting, are more likely to bring in ticket sales. 

“This new model offers the potential to alter the way in which we make decisions for the better, to be more quantitative and dispassionate to make better-informed choices,” vom Saal says.

In addition, Symphony executives believe the tool, when refined, could be used to help other symphonies, and could someday be commercialized to sell to other performing arts venues. 

Regression analysis is a mathematical method of determining which factors, or variables, are most important by tracing the link between each variable’s impact and the percentage of seats sold at the Fox. 

Weiling Zhu, chief financial officer for the Spokane Symphony, says she and vom Saal have been considering a new approach to booking for about five years. Previously, deciding which artists to bring to the theater was largely driven by gut feelings, Zhu says.

In addition to the Symphony’s 65-some performances annually, the Fox Theater, at 1001 W. Sprague, hosts notable performers in music and comedy, including Bryan Adams, Steve Martin, Old Crow Medicine Show, and Weird Al Yankovic, Zhu says.

COVID-19 took a toll when live events were shut down. Although some shows—like Weird Al—sell out, the Symphony found ticket sales to be volatile even after the theater reopened last year, Zhu says. 

However, well before the pandemic, the Symphony had struggled with the uncertainties of booking performance artists.

“In some circumstances, we thought it was (enough to use) our judgement to say, ‘we believe this is going to sell well,’ and it didn’t. We ended up with a loss,” Zhu says.

She says performing arts events venues rely heavily on ticket sales, but typically have little to no control over ticket prices.

“When we sign a contract with an artist, they usually have a (fee) scale,” Zhu says. “There’s some room that we can work with, but, in general, not a whole lot.”

While the Fox Theater receives donations, Zhu says about half of its operating income is tied to ticket sales and the number of performances sold. The symphony brings in about $4.5 million in annual revenue, Zhu says.  

Zhu says technically, the Fox Theater has never been in the black, partly due to volatile ticket sales and partly due to the depreciating value of the theater building. 

“The Fox operation ... is driven by how many concerts that we can do,” Zhu says. “The net profit, compared to the gross revenue, is very different.”

An artist can generate a lot of gross income in ticket sales, but Zhu says that after subtracting artists’ fees, the venue can be left with comparatively little net income. Meanwhile, a less popular artist who charges lower fees could result in higher profit margins for the Fox. 

This year, vom Saal and Zhu decided to pursue a strategy that relies on data. They approached Zhu’s daughter, Jeslyn Cai, a high school student at Lewis and Clark High School who has taken advanced-placement statistics classes, about creating the new data-based method.

Cai volunteered to tackle the project over the summer, with some guidance from Danielle Xu, professor of finance at Gonzaga University. Cai compiled data from several sources, including sales data from industry publication Pollstar, social media consumption data from music data analytics company Chartmetric, and demographic data from the U.S. Census Bureau. She examined more than 3,000 concerts by 17 artists from 1999 to 2022.

Zhu says the data includes album sales in the Spokane area, streaming of an artists’ work through platforms such as Spotify, as well as proprietary data on artists that have previously performed at the Fox. Each data set is treated as a separate element, or independent variable.

“We also look at economic numbers, like the area’s median household income,” Zhu says. “Mathematically, the more independent variables you look at, the higher your predicting power. We look at 27 independent variables.”

After compiling data, Cai began using regression analysis to develop a forecast model that ultimately predicts the percentage of the Fox’s 1,600 seats that will sell for a given artist’s performance.

Zhu says the success of an artist at the Fox is measured in the percentage of seats sold for a couple of reasons. The gross revenue for a performance or the number of tickets sold aren’t necessarily strong indicators of high local audience demand, she says, and using a percentage helps more easily compare the Fox to venues of different sizes in other cities.

“The first artist turned out to be one that we decided not to book, because the ticket sales didn’t come out to how we wanted them to be,” Zhu says. 

Vom Saal says, “We began getting a picture of what data seems to matter more.”

He gives an example of an artist who had previously performed at the Fox on a Saturday evening.

“We know what they sold on a Saturday night. Now, they want to come through, but they’re coming on a Wednesday,” he says. “The question is, what are they worth on a Wednesday versus a Saturday? ”

Cai, vom Saal, and Xu collaborated on a white paper reporting their methods, called “Predicting Symphony Ticket Sales—A Case Study,” which was released by the Symphony in October.

Zhu says the white paper could help other venues and performance organizations in a similar position to that of the Symphony. 

“It’s a game-changer in a nonprofit world, especially for an organization that doesn’t have a lot of resources,” Zhu says. 

Large national performing arts events venue operators, such as Ticketmaster, already have created tools similar to the Symphony’s regression approach, Zhu says.

“But most local nonprofits either don’t understand the dynamic behind it, or they just don’t have the resources to do it,” Zhu says. 

Xu says the model the Symphony worked up is innovative for the smaller venues in the live performance industry.

“It can be easily replicable by other symphonies and performing arts theatres across the country,” Xu says. “The publication of this research will shed light on how to use data to make sales predictions and optimize revenues.”

Vom Saal says he’s not sure whether the Symphony could or will commercialize the regression approach.

“Maybe we’ll do something in coordination with a more digital-first company, or maybe we’ll do it with some other groups in a consortium downtown,” vom Saal says. “I think we’re going to continue working on it, getting our feet under us a little more, and getting more confidence that it actually delivers something of value.”

Vom Saal says he doesn’t necessarily believe the regression approach will revolutionize the industry quickly, but he hopes that it will help the Symphony to determine which performing artists are most likely to bring in large audiences.

“Given time and continued evolution and lots of incremental steps ... I could see this having some real palpable impacts for our organization,” he says.