PWC reports the global entertainment and media (E&M) industry is set to hit a staggering USD $2.2 tn by 2021, and it is currently going through a transformation process. It encompasses various fields including movie studios, marketing service providers, publishers, service media organisations.
The broadcasting and media sectors – The role of predictive analytics
Companies are asked to better predict demand for their content; to ensure they give viewers what they want, when they want it. To succeed, businesses need to build a culture infused with analytics, adopt a curiosity driven approach, and apply creativity through the whole process. Creative data can improve accuracy, and that is the trick to predictive analytics challenges.
Why is it important?
Predictive analytics represents a huge opportunity for businesses. Organisations are trying to move from understanding what they have done to understanding their audiences at a whole new level, as well as to:
- Proactively marry the audience to content
- Leverage key insights into relevancies
- Adopt of new technologies
Successful marketers and business leaders need to be able to interpret quantitative and qualitative data, and determine where they can proactively deploy investment efforts. Marketers can successfully compete in a data-driven world by adopting a creative thinking that allows them to maximise benefits from digital channels, weave data fabrics, and draw businesses insights.
A real-world example and a taste of working in an international and multicultural environment
Last month, ESCP Europe’s Prof. Dr Hsin-Hsuan Meg Lee, Co-Academic Director of the MSc in Marketing & Creativity (MMK) set a one-week, multi-campus exercise in collaboration with Dr Lorena Blasco-Arcas, Academic Director of the MSc in Marketing & Digital Media (MDM), Dr Wei Zhou, Academic Director of the Master in Big Data and Business Analytics (MBD) and Dr Kamran Razmdoost, Interim Co-Director of the MSc in Marketing & Creativity. They asked 135 students based in Paris, Madrid and London to provide a forecast of what will happen in the Eurovision Song Contest.
2018 Eurovision Song Contest
This year’s contest took place in Lisbon, with 43 countries taking part but only 26 going through to the grand final. Due to their financial contribution towards the competition, the “big five” (the UK, France, Germany, Italy and Spain) and host country Portugal automatically classified. All entrants competed at two semi-finals. The contest attracts 200 million viewers and is notorious for geopolitical voting bias. Performances are not necessarily judged by their entertainment value or professionalism. More information is available here.
Dr. Lee emphasised that, “As educators, we are always looking for ways to combine creativity and analytics. Usually, when talking about analytics, people tend to focus on rigorous scientific training, extended datasets and quantitative modelling. They are important, but we believe the real insights come from one's profound understanding of the domain, and the creativity in combining different quantitative and qualitative elements. This allows us to turn sometimes arbitrary values and observations into something useful.”
The teams had to:
- Identify key performance indicators
- Take data from a variety of sources, including social media
- Build the predictive models
- Predict the final ranking
- Choose a country and analyse its audience
Students had to look at different variables such as: historical data, social mentions, voting allies, performance vote, behaviour proxies (e.g. consumer expenditure on leisure activities, karaoke bars, imports of musical instruments), Google searches by country per year, people listening participant songs via different platforms (e.g. Spotify, YouTube), Hofstede cultural index. From this, they had to predict the Eurovision winner.
Participants received three awards including:
1. Best prediction award
2. Eurovision expert award
3. Most creative predictive model award
The learning experience
A clearly defined and collaborative strategy was crucial for this exercise. Key factors to develop the models were:
- Determining the variables to implement
- Robust social media view
- Strong communication skills and the ability to come up with a creative strategy
- Establishing deadlines and milestones; crucial due to the project’s complexity
- Strong quantitative and qualitative analysis. It’s not feasible to measure all aspects!
- Building shared intention purposes
Winning teams: What were the results?
Dr Kamran Razmdoost announced the winning teams at the end of the exercise.
- Best Prediction Award: Team 4
Members: Diana Arellano (MMK), Jay Kwok (MMK), Maria Paez, (MMK), Alizée Cadiou (MDM), Jose Manuel Lopez Carvajal (MDM), Moritz Barbarino (MBD) and Maximilian Stoeckl (MBD)
- Eurovision Expert Award: Team 10
Members: Jeremy Jackson (MMK), Alice Vigneron (MDM), Catherine Warde (MDM), Ivett Molnar (MDM), Chandrika Sharm (MDM), Mathilde Gelin (MBD) and Fabrice Zaumseil (MBD)
- Most Creative Predictive Model Award - Team 13
Members: Clara Corsin (MMK), Jose Mario Ocampo (MDM), Laurene Mathieu (MDM), Simon Szlper (MDM), Sophia El Akroud (MDM), Anke Joubert (MBD) and Ankita Naik (MBD).
Dr Kamran Razmdoost revealed that: “As expected, the success of teams depended on the way they could bring both creativity and analytics into one place. Team 13, as the winner of the most creative model, looked beyond obvious factors determining the winners and high performers. For instance, they used the number of Karaoke bars in the capital of each country to investigate the level of social engagement with the music entertainment industry. In fact, most of the time, we do not have the data that we like and we need to create ways to reach the insights good enough to support our judgements.
Also, creativity contributes to analytics by helping marketers and business experts define what matters to them. Team 4, the winner of Best Prediction Award, focused on the artistic quality of the performances, analysing them based on three important aspects, and applying analytics in learning how these factors determine the success of performers.”
Making creative predictive analytics work – Proven keys to success
- Define the problem
- Determine the key variables implementing creativity through the whole process
- Identify the right data that supports the variables
- Analyse and enrich the data
- Build models for predictive analysis
- Experiment the model and implement the analytics
Jose Mario Ocampo, current MDM student explained that, “During the different stages of the contest, we focused on facts for the last 10 years, the more factual the better.” The model built predicted that the top five will be Israel, Italy, Sweden, Cyprus and Spain. The overall exercise was a great success as the actual results consisted of Israel, Cyprus, Austria, Germany and Italy in the top five positions.
If marketers and pioneering thinkers are to overcome the issues that arise in the digital age, they must be challenged to successfully merge creativity, marketing, technology and management. This real-world exercise equipped students with the tools required to deal with a full range of businesses outcomes, and build a successful career in today’s business environment.
ESCP Europe and its CMC would like to congratulate the winning teams on their outstanding performance.