Big Data for Music Festivals

Using insights gathered from data to improve the design, planning and sustainability efforts of future festivals

 

The Roskilde Festival is the largest culture and music festival in Northern Europe, attracting over 130,000 visitors annually (2015). Over the ten day festival, more than 170 live music and cultural acts perform and festival-goers consumes over 200 tons of food and generates over 300 tons of waste.

In 2015, Copenhagen Business School partnered with IBM to create a big data laboratory during the Roskilde Festival. They wanted to use the festival as test bed for research into sustainability efforts and issues of urban environments. The festival was chosen to as it was seen as a microcosm of a complex modern city. The goal was to gather data in this real-world laboratory to devise and test initiatives to improve sustainability efforts.

For the festival producer, gathering this data and insights can help improve the sustainability of the festival and provide potential cost reduction benefits. On the revenue side, the insights could lead to better attendee experiences and drive more returning consumers: reducing churn and thereby increasing the lifetime value of festival-goers.

 

The project team collected a massive amount of data during the 2015 festival, amounting to over 91 million tracking points. These were unstructured data, primarily collected from festival-goers’ smartphones through an opt-in program. Examples of metrics captured are: movement of crowds through the event, sales of food and beverages over time, popularity of different musical acts, weather data. They also conducted over 12,000 interviews with festival attendees. The team used cloud-based analytics, which was much faster to set up, manage and eventually scale.

These collected data was used to investigate relationships and interesting correlations between these metrics. Some of the questions asked were:

  • Relationship between unique users and total minutes watched of event X
  • How many people visited each concert and how long they stayed there
  • Which concerts were watched by the same people
  • Which food courts and bars were most popular
  • How many people shopped outside of the site during the festival

The team used this data to help construct the flow of how people travel in the site over time to enhance planning and designing a safer and better experience for festival-goers. They found trends of food sales with respect to temperature and time of day to help merchants strategize and plan their inventory, contributing to reduced waste and added profitability. This data was also used to gather insights into personality types of consumers. An ongoing project is underway to uncover and understand behavior insights of people who watch certain combination of artists, like what type of food or brands they may prefer. These data could also help the festival producer better understand factors that lead to customer churn, customize and boost their marketing efforts to the different consumer types to drive revenues.

Image result for roskilde festival

 

Looking forward, the festival producers hope to develop predictive models for these patterns and trends. The aim for the following year’s festival is to execute on real-time analytics during the festival, producing live updates and recommendations for participants to enhance their festival experience and to alleviate congestion and risk of accidents. They aim to provide live information such as how crowded each concert is, lines for food outlets and bathrooms to help with crowd control. The longer term goal of these efforts is to use insights gathered from data across multiple years to improve the design, planning and sustainability efforts of future festivals.

 

Sources:

https://channels.theinnovationenterprise.com/articles/predictive-modeling-around-music-festivals

https://www-01.ibm.com/common/ssi/cgi-bin/ssialias?subtype=AB&infotype=PM&htmlfid=CDC12349USEN&attachment=CDC12349USEN.PDF

 

Previous:

Amino: Taking the Guesswork out of Healthcare

Next:

Broadway Enters the ‘Big Data’ Era

2 thoughts on “Big Data for Music Festivals

  1. Interesting post Ophelia, I’d never heard of the Roskilde Festival before! I thought of this post since Coachella is going on right now and looked up attendance from 2016: apparently there were 99,000 people per-day in attendance during the six days of Coachella. There’s a pretty large opportunity for Goldenvoice, the concert promoter, to use data and analytics to improve the experience at Coachella and Stagecoach but I haven’t been able to find much on whether they are partnering with any companies like Roskilde did. Do you happen to know if Goldenvoice is indeed doing anything similar? I found an interesting article on the economic impact of Coachella 2016 here in case you are interested: http://www.latimes.com/entertainment/envelope/cotown/la-et-ct-coachella-economy-by-the-numbers-20160420-story.html

    1. I believe Coachella and other large US festival producers are also experimenting with new technologies that help collect data for their festivals. There are RFID wristbands and other wearables that some of these festivals were trying out. This can help the flow of traffic (operational ease), but also help producers track traffic and improve on festival design. I believe the Roskilde festival study has been the most in depth so far, partly because it was an academically driven study, and the only one with the explicit interest in sustainability efforts. Hopefully other festival producers will also pick up on this and challenge their current way of operating!

Leave a comment