Data Technology Fan Activation

The New Rules Of Fan Engagement

By David Longstreet on September, 27 2018
David Longstreet

David has over 20 years of building analytical research, statistical and econometric models. He leads FanThreeSixty’s core fan analytics capabilities, which includes predictive sales and attendance models, enriched fan profiles and smarter audience segments.

What one fan does is not as interesting as what 10,000 fans do. This is why the role of data science in sports goes beyond player performance. When internal teams have an organized database of fan data, it can be mined to uncover distinct attributes and behaviors that make up each unique fan. Machine Learning algorithms can also be used to find unexpected patterns of fan behavior, leading you to trends or insights that can change how you engage with your fans. 

Machine Learning algorithms can also be used to find unexpected patterns of fan behavior, leading you to trends or insights that can change how you engage with your fans. 

Knowledge
Until very recently, most fan "learning" or knowledge was based on experimentation and observation. It was a manual, lengthy process. We now live in a hyper-connected world where technology is ever-evolving, which is fueling a newer, faster way to learn: Machine Learning (ML). ML removes the manual labor and reduces human error. It allows us to string together massive amounts of data using computer-based algorithms to form knowledge in an automated fashion. These algorithms then uncover underlying patterns in fan behaviors that are hidden in the data—in seconds.

Machine Learning removes the manual labor and reduces human error. It allows us to string together massive amounts of data using computer-based algorithms to form knowledge in an automated fashion.

So What Does it Mean?
We’re often asked by our clients, “What exactly is machine learning?” Well, imagine teaching a computer to play chess. You can do it one of two ways: 1) researching and analyzing ALL the rules to the program each play and the way each piece moves individually or 2) use data science to develop a series of algorithms that analyze all possible chess moves from millions of chess games. In the latter scenario, the computer begins to quickly understand and “learn” the rules of chess based on patterns and movements without being explicitly programmed. Rather, the machine (computer) automatically learns from the data, finding patterns that allow the machine to make better decisions and improve from experience.

...the machine (computer) automatically learns from the data, finding patterns that allow the machine to make better decisions and improve from experience.

Bringing Big Data to Fan Engagement
Now imagine taking a decade’s worth of ticketing data, supplemented with several other inputs: actual attendance data; win and loss records of the team(s); importance of game; weather data; fan demographics; loyalty programs; season ticket renewal information; price of tickets; concession spending; parking fees; merchandise purchases; competing events in the city; and so on and so forth. Now imagine repeating this for every team across an entire league. The result would be a database of fan patterns and behaviors based on hundreds of event variables and millions of historical observations about fans.

The result would be a database of fan patterns and behaviors based on hundreds of event variables and millions of historical observations about fans.

Predicting Fan Behaviors
Once the above database is created, a data scientist can write algorithms so the machine (the computer) can learn fan patterns that exist within the data. So like the chess example, the computer first learns the distinct patterns of fan behaviors that then make up the rules of fan engagement. From this, predictive models can be built to help marketing, sales and operations teams make better decisions and improve from past experience.

From this, predictive models can be built to help marketing, sales and operations teams make better decisions and improve from past experience.

For example, this information can be used to understand which games may be at risk for low attendance or uncover the best time to re-target a first-time ticket buyer with a discount to buy again. It can also be used to alert ticket managers when season ticket holders are at risk of not renewing. For example, at FanThreeSixty, we built our proprietary renewal model from variables like attendance history, demographic data and more, to change how ticket managers act on data. They are now proactive, engaging and capturing these at-risk fans before he or she decides to not renew.

So whether you’re a chief marketer, a sales manager or an operations director, apply the power of big data to your fan engagement strategies to become both predictive and prescriptive.

...apply the power of big data to your fan engagement strategies to become both predictive and prescriptive.

It will make it appear as if you’re playing chess while everyone else is playing checkers; and in this instance, you don’t have to know all the rules—after all, that’s why the machine exists.

Stay up to date