Jun 10, 2021

Identifying Returning Moviegoers With Machine Learning

An important piece of every exhibitor’s reopening plan will be the identification of those moviegoers most likely to visit, in order to target them with offers and information that ensures that they return and enjoy the cinema experience. As a number of our exhibitor clients have been operating for the past 6-9 months, we wanted to find out whether it is possible to use pre-pandemic visitation habits to identify those moviegoers most likely to come to the cinemas during (and after) the pandemic. 

Fortunately, our data science team discovered that this is in fact possible, and in the process also noticed a few other interesting trends. Our initial analysis took into account data from a number of exhibitors in the US and we then extended it to the United Arab Emirates (UAE) to validate that the results hold in other markets as well. You can find the full in depth report in our latest data science blog, or an overview of the key trends we noticed below:

  1. The data confirms that expected factors such as pre-pandemic visit frequency and age are highly predictive of whether a moviegoer will return i.e., those with a higher visit frequency prior to the pandemic, are more likely to come back first.
  2. Moviegoers that spend more are more likely to visit after COVID-19 - suggesting that the returning audience is being driven by families / groups, regardless of movie genre.
  3. Loyalty engagement including data sharing (valid mobile/email, date of birth), and in some cases email interaction (opens/clicks), predicts the likelihood of visiting the cinema throughout the pandemic. What this means is that those moviegoers that have the highest probability of coming back, usually are the ones that have shared a valid phone number, have shared their date of birth, and regularly interact with marketing emails and campaigns.
  4. We observed mostly the same picture in every exhibitor we looked at, including the US and overseas (UAE).

In addition we also decided to examine more closely some trends around recent titles, including Mortal Kombat (Fantasy, Action, Adventure), Godzilla vs. Kong (Science Fiction, Action, Drama), and Tom and Jerry (Comedy, Family, Animation). These titles represent a large share of the box office in 2021 so far, as part of the recovery of the exhibition business in the post-COVID era. Analysing data from a US Exhibitor, we found a familiar picture, where the attendance likelihood for these three movies is being driven quite strongly by whether moviegoers have submitted valid personal data as part of the loyalty program (mobile number, age), frequency of visitation prior to the pandemic, spend, and the engagement with marketing campaigns (clicking email links).

If a factor is important to predict whether a moviegoer is likely to visit, you would expect to see either some blue or red, as far as possible from zero. If the coloured dots cluster around zero, then the related factor doesn't seem to be as important.

These insights are consistent with the results above, and suggest that these trends generalise across movie genres. Another insight of interest suggested by our results is that, in the context of these movies, Baby Boomers are less likely to visit when compared to other demographics (Millennials and Zoomers). This is likely due to increased caution given their higher risk of adverse outcomes associated with COVID-19.

Altogether, this was an interesting project, and a nice example of how Machine Learning can be used to do precision marketing. Using AI and Movio data, we were able to build a segmentation tool to be used to help our customers in the exhibition industry to identify those who are more likely to visit the cinema as soon as it's safe to do so. We were also able to uncover interesting insights that will inform the marketing strategies of the exhibitors we work with around the world.

If you’d like to continue reading, you can see the full data science version of this blog here. To learn more about our new segmentation tool or how we can help with industry insights please reach out.

Written by

Dr William Caicedo

Connect with us