INTRODUCTION

Those were very different days when we had to wait for blockbuster films to be released on large screens or for a week to catch up on the latest episode of our favorite tv programs. Online screening and video content platforms, also known as OTT (Over-The-Top) services platforms, have successfully established their presence in the entertainment sector and people’s lives. Personalization and customization of products or services are significant features of this new digital world. Netflix is one of such OTT platform.

Netflix is one of the world’s leading entertainment services with 221 million paid memberships in over 190 countries enjoying TV series, documentaries, feature films and mobile games across a wide variety of genres and languages. Since entering the streaming market in 2007, Netflix’s growth appears to have never slowed. They have over 8,000 movie and TV series on their platform for over 200 million paid subscribers as of mid-2020. It’s hard to believe that Netflix was once a tiny firm focused on DVD rentals since 1997. Netflix’s commercial success is based on its massive database and detailed records, which support its advanced recommendation algorithms. This project is a visualization which will include three areas of data on Netflix: geographic, time based and popularity.

INSPIRATION

I look for inspiration related to the dataset and the topic. In this project I wanted to create advance visualization. So my focus was on searching for inspiration that will challenge me and help me to create my visualization. I created an inspiration board as I was inspired by multiple work. The inspiration board involve four projects: The air we breathe, How many films have I watched on Netflix, Netflix data visualization and Filming in NYC.

MATERIAL & PROCESS

Tools 

OpenRefine – A tool for working with messy data. 

Tableau Public – A data visualization software. 

Kaggle – open data resource

Design Process

Searching and Cleaning up the data

Initially, I decided to work on a particular dataset that has 4 OTT platforms and their comparisons. That data was not clean and I couldn’t clean it without coding. So after struggling for 3-4 days I changed it and recentred my focus on a single platform: Netflix. The second data on Netflix was also not clean. So I cleaned that data on open refine. I split countries into different cells so that tableau can easily detect those countries and give the output. I also added two other datasets with the main one to add more details to the final visualization. All the datasets were: Movies and TV Shows on Netflix, Netflix subscription fees in different countries, and Netflix subscribers.

Brainstorming

The third and important step after searching and cleaning the data was to think about how it will look and what all features it will have. I did some brainstorming before directly jumping to the software. Below are some of the samples of my brainstorming process

Color Palette

Creating Visualization

With the previous projects, I learned the fundamentals of Tableau. However, in this project, I hoped to learn and implement some advanced charts in my visualization. In addition, I wanted to present my final visualization as a story. I watched a ton of tutorial videos to learn advanced calculations and charts.  

1. Geographic Impact:

The first dashboard’s data focused on Netflix’s geographic impact. First, I made a Sankey chart with a map (tutorial: https://youtu.be/iIr8bcOlnIk) to show which countries have what kind (Movies/TV shows) of content and how much of it they have. I also included a dataset with the costs of plans available on Netflix. To make it more interactive and informative, filters were added. It was difficult to create this Sankey chart, and I’m not sure where I went wrong. Since it didn’t turn out as expected. I wanted to get rid of the extra lines on the map that showed the countries’ data.

2. Timeline:

The second dashboard was created with the purpose of showcasing the growth of Netflix over the years from the very beginning. This dashboard was also made interactive by adding filters to it and creating a line graph to. show the growth easily. The initial idea of this was to create an animation but the output I wanted was not possible on tableau. I crated the radial chart (tutorial: https://www.youtube.com/watch?v=ntysZPxchm0 ). The Initial idea of the chart was a little different by after playing with the size a placement of dots I finalized the below one.

3. Popularity

The third one was based on showcasing the popularity of Netflix worldwide. The emphasis here is on displaying the number of subscribers as it varies by country and year. This was not an advanced graph, but I believe it accurately represented the data, so I used it. Because there isn’t a lot of information and it’s easy to understand, I didn’t add any filters.

4. Homepage

I wanted to design a dashboard where users could easily find information and navigate. The goal was to make it visually appealing while also avoiding the need for additional navigation instructions. I came across a data visualization by Linus Tse while researching and looking through a bunch of visualization on the same topic, and I like the concept. So I made a homepage that resembled the original application’s user interface.

UX RESEARCH

I conducted user experience testing on the drafts with two participants to understand how they understood the visualizations, whether there were any areas of confusion, and what they took away from the graphics. I selected two participants who matched my target demographic. Both were millennials between the ages of 25 and 30 who live in the United States.  Most importantly, they had Netflix subscriptions and are frequent users. I conducted user tests via video call on Facetime, with participants sharing their screens so I could see how they engaged with the visualizations while taking notes. I used the think-aloud technique.

PROFESSOR AND PEERS FEEDBACK

The final feedback I received during the presentation was to make few changes with label in the dashboard 3 and color change in dashboard 1. I did the changes as shown below. Also, they like the homepage.

RESULTS

Overall, the visualization was successful in conveying significant Netflix information through the use of a story. The dashboards made that quite clear.  The United States has the most subscribers and content in the whole world. The United States is at the top in terms of subscribers, but if we look at subscription growth, Africa is at the top. It turned out that Switzerland is where Netflix is most expensive. By clicking on the homepage image below, you can navigate to the final result.

REFLECTION

This project was the most exciting and challenging of all the projects I’ve worked on so far. I pushed myself and got out of my comfort zone to learn advanced Tableau features. I spent a lot of time just watching and learning how and what to do. Although I only had a short time to learn everything, I gave it my all. The most difficult aspect for me was completing this project in such a short period because I wasted nearly three days of my initial phase time trying to figure out and clean the initial dataset that I proposed. After switching that data set, I discovered three good datasets. It would be ideal if the subscribers’ data set contained more years of data. We could add more information in the future, such as IMB ratings and data on how many people share a single subscription. It’s great that the homepage interface allows you to add more information without changing much.