Building a fanbase

In this first project, I'm going to analyze the film industry using datasets from IMDb.com. The goal is to explore the data and help Microsoft's Team to understand the market and elaborate a strategy to enter this industry.

Why looking only at IMDb data?

From all provided datasets, IMDB.com was the only source that had several files and a lot of information from different levels. I could try to merge all different sources but since they don't share the same title_id it would be difficult to merge using movie_title. So for simplicity, I'm limiting my analysis to one source.

Caveat about IMDb's datasets: It seems like the files were extracted in 2019 based on the No. of Movies over time. Therefore, more work could be done on getting updated data.

Navigation

You can find the following files/folders in my repository:

Film Industry Analysis.ipynb

• Content: Main Jupyter Notebook with the whole analysis.

• What I'm doing:

  1. Unzipping files and exporting them to a different folder.
  2. Exploring the available datasets, understanding what's available, and the structure/granularity/pitfalls of each dataset.
  3. Asking questions that will help Microsoft's team understand the industry, then answering them using datasets and developing recommendations based on my findings.
  4. Presenting results/conclusions through visualizations.

presentation.pdf

• Content: Non-technical presentation to stakeholders talking about the approach and recommendations.

zippedData/unzippedData folders

• Content:
  • Zipped/unzipped files from the following sources:
    • Box Office Mojo
    • IMDB
    • Rotten Tomatoes
    • TheMovieDB.org
  • Context: Those are well-known movie websites with a lot of information related to movies, tv-shows, video games, and streaming content - including cast, ratings, movie length, titles in different places, budget, trivia, goofs, rankings, personal biographies, plot summaries, and fan and critical reviews.

Analysis Approach

The datasets provide a lot of useful information. However, to develop a strategy to enter the film industry, I'd like to find trends in successful movies, which means that first I need to define success.

Definition of success and how to measure it

Success can be defined in many ways, but I suppose that Microsoft would like to have popular movies with high views, ratings, and maybe profits. It's hard to expect that their first venture will be popular, let alone profitable.

Consequently, I will measure success based on the Number of Votes and Ratings. Unfortunately, I'm limited to what's available, and the number of votes might not be an ideal proxy for views, but it's the closest I can get.

From my perspective, Number of Votes can be even more valuable because differently from a view where it's just one person watching a movie, a vote means that someone went to IMDb's website, created a profile/logged in, searched for the film and voted. Since we have all these steps to complete to vote, I'd say that votes are more meaningful.

Steps to take

Given the business problem, I'm analyzing the Industry taking the following steps:

1. Validate that it's a growing industry: First, I need to understand whether it makes sense to enter the Film Industry. Therefore, I'm going to analyze the market and see if it shows signs of health. For that, I'm studying whether the No. Released of Movies is growing every year. The Demand and Supply Rule says that supply will follow demand. Therefore, if I confirm that the number of movies is going up every year, most likely, the same is happening with the number of customers/viewers, and, consequently, it's an investable market.
2. Understand which styles/genres are growing: The next step is to dig a bit more and analyze the same number by genre. I'm following the same Demand/Supply Rule, trying to find which genres are more popular. With that, we can identify opportunities or avoid creating a product that would face a lot of competition.
3. Understand what customers are telling us: After looking at the No. of Released movies, it's time to look at the data points that will tell me what customers want. For that, I'm going to analyze which movies have High Ratings and Votes. The No. of Votes tells a different story here. If someone is willing to go to IMDb.com and voting, it's because they hated/loved the movie or love movies in general (that's my hypothesis). Either way, it's a different perspective, and by taking this approach, maybe I can find out what customers are trying to tell. I understand that this approach's problem is that it will ultimately be skewed by the most voted movies out there. Also, IMDb.com represents only a fraction of all customers, maybe not representing the whole population. So any recommendations/conclusions would require more research (just like any conclusions here).
4. See if specific roles impact No. of Votes: This might be a long shot since I don't have all the data for all movies because the dataset I'm working with shows only the 'principal' roles. But the idea is to see if certain roles have more power than others in terms of the Number of Votes they bring. With this, Microsoft can dedicate their time/money to roles that matter the most because you cannot have the best on everything in a movie.
5. Impact of releasing a movie internationally: Finally, I'm trying to find if there's some correlation between the No. of Votes and No. of Countries/Regions in which the movie was released. With this information, Microsoft knows whether they need to invest money in making the movie famous in other countries or not. Making a movie an international success is expensive (Marketing Materials, hiring other actors to translate the film in different languages, organizing launches in different countries, etc.).

Questions

Below you can find the questions I raised during the project. I'll also add an explanation for why I think those are valid and important.

1. Is the number of new movies growing over time?

Reason: We must understand whether this is a market that is producing more and more content over time.

If Microsoft is thinking about entering the Film Industry, this question will give a good understanding of whether it's a growing industry and if there's a lot of competition. It's expected that if the supply of movies is increasing, the demand is going in the same direction, so it's probably a good market to enter.

2. Which genres have the most movies?

Reason: By answering this question, I can understand a bit more about the demand. I believe that producers would not create movies of genres that are not popular.

Having this information helps Microsoft to understand the market and start thinking about which strategy to apply (compete with the popular genres or focus on others/not so saturated genres).

3. Which genres have consistently had the most No. of movies over time?

Reason: With this question, I can find trends of genres that are becoming more popular over time and maybe suggest Microsoft follow the same trends.

4. What are the best/most rated genres?

Reason: So far, I have done most of my analysis using No. of Movies. This time, I'm going to use No. of Votes.

I think that Ratings/Number of Votes can be more meaningful because the act of voting requires someone to go to IMDB.com, search for the movie, and vote. Therefore, I'd say that it gives us a good understanding of the market from the consumers' standpoint. Also, if someone voted, it's because they watched the film, so the number carries more meaning.

Therefore, by asking this question, we get a different perspective on genres and a good idea of which one would have a higher chance of success.

5. Which roles are the most important in a movie?

Reason: The idea of this question is to understand which roles require more attention or are more likely to bring more views/votes to movies. Maybe there are some similarities among successful films.

By knowing which roles bring more votes, Microsoft can plan their budget accordingly or dedicate more time to find the best on important roles.

6. What's the impact in terms of the popularity of having a movie playing internationally?

Reason: This question aims to understand whether Microsoft should also focus on making the movie famous internationally or if most of the popular films didn't necessarily were played in different countries.

By launching a film in different countries, there's much preparation and other things to consider (Marketing Material, Subtitles, Local Titles, etc.). So knowing this ahead of time is helpful.

Conclusions

1. The Film Industry is a competitive and growing market: The supply of new movies is increasing over the years. The reason for that can be the following:

• Movies have become cheaper to make over time as technology became more accessible to anybody.
• The market and customer base grew. So the demand for more content is growing.
• People have access to the internet + Consumers are moving from Cable TV to On-Demand subscriptions.

Those are all positive reasons for Microsoft to enter the Film Industry. The only problem is that competition has grown a lot, and Microsoft will compete with Apple, Netflix, Disney, Amazon, and YouTube. Moreover, there's a lot of discussions on the subscription market's saturation, so their pricing will have to be competitive.

2. Documentary, Drama, and Comedy are the most popular genres in No. of Movies: After analyzing movies in IMDb's database, it's possible to see significant numbers of Documentaries, Drama and Comedy movies. When looking at different combinations of genres, Comedy + Drama and Comedy + Drama + Romance have had the most No. of Movies over the last ten years. For Microsoft, these might be too saturated genres to enter but show a picture of the industry.

3. Building a Fanbase: During the analysis of the Most Voted/Best Rated movies, Action, Adventure, Sci-Fi, Adventure, Animation, Comedy and Action, Adventure, Fantasy showed up a lot over the last ten years. Those were the genres with the most No. of Movies in the Top 1% of movies in terms of number of votes.
I think the reason is that most of these movies are superhero movies (Captain America, Iron Man 2, Thor, and Wonder Woman), part of a movie series (Avengers, Star Wars, Pirates of the Caribbean, Toy Story, Despicable Me) or Pixar Movies.
They attract a lot of fans over time. So once you have that fanbase, it becomes easier to create movies with a lot of views, votes and become popular.
Therefore, a good strategy for Microsoft might be to enter this world and try to partner with another company like DC/Marvel and create an exclusive superhero series. The good thing about a partnership is that they wouldn't have to develop a new story since they would use an existing character.
Also, another option would be to develop a new Adventure, Animation, Comedy story with unique characters like Minions on Despicable Me. This one is a bit harder to execute, though.

4. Hire the best on roles that matter the most: After analyzing which roles were the most frequent in high voted movies. The recommendation is to focus on getting the best actors, then writers, producers, actresses, and directors. Note: Unfortunately, this analysis is biased towards men since actors have higher importance relative to actresses. And the reason is that there are more actors in high voted movies. I'd probably group both as the same to avoid adding this type of bias.

5. Have an international footprint: The Average No. of Votes increases as the No. of Regions in which the movie is played increases. Therefore, my recommendation for Microsoft would be to prepare to launch the film internationally.

Final Thoughts

More could be done to gather more insights. For example, I'd like to see the number of Tickets Sold, viewers + demographics, and more information on which Platform/Studio created the movie. Moreover, understanding Microsoft's goals would also guide my analysis to relevant conclusions for them. For example, I don't know if they're trying to be profitable in this endeavor or if the goal is to enter this industry and then worry about profitability.