From renowned sequels to niche art films, the range of new movies that come out every year is diverse and so are the audiences that come out to see those movies. With every release, cinemas have to ask themselves: is this going to be a box-office hit or miss? And more importantly: how many people can I expect to turn up for these movies and how can we maximise sold tickets?
It’s one thing to expect a certain attendance rate for a particular movie, but does that expectation also hold true for every cinema? Some movies might attract bigger audiences for a certain cinema than another simply because that city tends to like a certain actor/director/production/you name it. So how can cinemas find out if they have a Brad Pitt bias and what impact it has on attendance rate? Perhaps the most reliable answer lies in the insights that can be dug up using AI methods.
There are many factors that contribute to a movie’s success, like cast and directors. If we all knew all the success factors, every movie could be a hit. That doesn’t mean it’s impossible to come close to good predictions. An educated calculation is always better than a gut-feeling best guess.
In order to make that calculation, we need data. Cinemas typically don’t have access to promotion budget of a movie so trying to project movie popularity based on how much money is spent on ads and PR is an entirely different practise. Cinemas will have to rely on public data, which is widely available. Think about IMDB ratings, genre, release dates, cast and so forth. With machine learning, ratings, and casting data from similar movies can be analysed to uncover patterns in movie popularity that will drive future attendance. By combining these trends with information about theater capacity, historical ticket sales, and showtimes, an algorithm can forecast how many tickets a future screening can sell.
Now that we know what the likelihood is of success for a given movie, the following question should pop up: “Is that also the case for cinema X”? The golden bullet here lies in combining movie popularity forecasts with cinema-specific data. The latter means that we take data that is specific to a specific location to forecast cinema attendance, including historical attendance rates. We can combine historical data from screenings of previous, similar movies, the popularity forecast of the movie of interest, and the projected screenings of said movie to forecast future ticket sales.
Once know how many tickets you’re going to sell based on the movie itself and the cinema capacity, it’s time to adequately schedule your employees to make sure customers are short of nothing. Knowing their demand allows cinemas to know how many cashiers, ushers, and managers you’ll need exactly when and where depending on your service levels. Cinemas that don’t mind longer queues can play around with shifts and the impact it has on the service level and costs. More importantly, demand is the perfect base to create and assign shifts to employees. As cinemas are typically characterised by higher demand in the evenings and weekends, different laws and rules apply when it comes to scheduling. And how about coordinating preferences of your employees? We can imagine that some of your employees prefer not to work on Saturday night, while others don’t mind at all.
All the decisions involved in making fair and cost-efficient schedules can be hard to make, especially when the demand fluctuates a lot from day to day, from season to season, from movie to movie. Workforce Optimisation solutions can help cinemas streamline their scheduling and get the most out of the in-theatre experience of their guests.
If you’d like to learn more about Widget Brain’s workforce optimisation services, go to https://widgetbrain.com/cinema-employee-scheduling/ or schedule a demo below to learn how you can make better attendance forecasts.
Suddenly there is a new reality. To handle change in unprecedented times is to learn and decide faster than before. That means old data has to be forgotten and new decisions have to be made in the near and far future. Read this full guide on knowledge, data-driven decisions and automation to help you fully prepare for the next disruption.
Due to the crisis, most generic demand forecasting models in place today are no longer as accurate as they used to be and a relative approach has to be taken in order to find the “new normal” when compared to traditional historical patterns. Read more about the new normal in demand forecasting.
As the challenges presented by COVID-19 continue to affect the workforce around the world, companies are finding new ways to prepare for the next unexpected crisis. Fortunately, various tech solutions provide the opportunity to do just that. In our eyes, one of the most important is automated scheduling. Read more.