How Will Automation Affect Your Job?

Role (Individual Work)
Three Months
Data Visualization
Adobe Illustrator

Over the last century, machines have replaced workers in many tasks. Fears of robot-induced unemployment often dominate discussions over the future of work.

This project aims to answer these questions:

  • Facing the challenges posed by automation, which occupations will thrive, and which ones will wither?

  • What the environment for career in the future will be?

  • How can we prepare for the change?


Data Collection

Collecting data was one of the biggest challenges for this project. In the beginning, barely could I find any data about how automation will affect jobs in the future rather than narratives. After weeks of searching for sources, I finally found two reports from Mckinsey Global Institue: Jobs Lost, Jobs Gained: Workforce Transitions in a Time of Automation and

Skill Shift Automation and The Future of the WorkforceAnother resource for my project was the report Automation and Artificial Intelligence: How machines are affecting people and places by Mark Muro, Robert Maxim, and Jacob Whiton.

Reading through the three reports which include over 300 pages in total was overwhelming but also full of those "A-Ha" moments. However, challenges never stopped. When I was reading the website for the report Automation and Artificial Intelligence: How machines are affecting people and places, I ran into the data I had always been wanting: the automation potential for specific jobs from different categories, real data points! As much I was excited, soon I was disappointed by the fact that there was no original data file appended to the report.

"Should I give up?"I asked myself.

"NO, never."

Two hours after that, I found a tool to scratch data from the website: ParseHub. The process of scratching data using ParseHub was fun but extremely time-consuming - as it became really slow when loading too many datapoints. I spent a whole week figuring out the syntax of the coding language in ParseHub and was finally able to download all the data of 21 categories of jobs!

Screen Shot 2019-12-12 at 9.42.39 AM.png
Screen Shot 2019-12-12 at 9.42.27 AM.png

Find the Story

As I dived deeper and deeper into this topic, I started to come up with the story that I wanted to tell. I would first start to talk about the current fear of automation potential of different jobs, and then show the historical trend of job shift when new technologies emerged where I would describe the lessons we learned from the history. After that, I would take the readers to look into the future: what are the new opportunities that could help the situation? In the end, I would like to discuss about how should we prepare for the change, from the perspectives of individuals, businesses, and governments.


How to make the story appealing?

In terms of design, I decided to use a little character representing the user and walk them through each chapter of the story. In order to create an immersive feeling, I used scrollytelling to separate the content into parts. The chart would change according to the content that is currently displayed to help with understanding. In addition, I used drawings to represent narrative data, for example: using leaves to represent trends that might contribute to the future career increment, using branches and leaves to depict that a promising future of humans required the efforts not only from individuals but also from business and government.

Given the fact that the data used in my story were half numerical and half narrative, the way I chose to represent them was critical to keep the story away from "being too dry and boring". The first thing that came across was how might I make people relate to the stories.


How to make the story appealing?

This project is very HEAVY on coding. The biggest gain from the project to me was a HUGE level up in D3.js, HTML and CSS skill sets. I scratched the surface of each of them a bit prior to the class. But now I can customize the chart the way that I want, make a static picture fading in/out, and change the status of different elements according to the chapter that the reader is reading, while NOT MESSING UP with other layers of graphics ("Otherwise you will have nightmares of debugging.").

I also learned about the thing that I always wanted to learn: scrollytelling. The scrollytelling function in my project was customized based on the library Waypoints.

There are still a lot of improvements that can be made. However, given the time constraints, I am so proud of my work.

Now, please allow me to introduce you the project "How Automation Will Affect Your Job?"