I discovered The Data School while at a crossroads in my career. My background spans multiple industries: I’ve worked in nonprofit operations focused on community change (should I explore NYC parks data?), I launched a podcast focused on consumer packaged goods (how many granola bars are out there—and are any of them actually healthy?), and I co-produce a Climate Week event focused on real-world solutions (what’s the state of carbon emissions in NYC buildings?).
I had no shortage of ideas—but I still didn’t know where to begin.
If you're anything like me and the open ended nature of picking ANY TOPIC feels a little overwhelming, here are some tips to demystify this open ended opportunity and get started on picking your first data set.
What Helped Me: Start With People, Not Data
I asked myself: Why do I want to be part of The Data School? The answer? I love helping people answer questions and fix things that aren’t working. As an ops person, I’ve often served as the bridge between people and systems—understanding needs, identifying bottlenecks, and building solutions.
So instead of diving into a dataset right away, I wrote down a list of three people doing awesome things—and booked a meeting with each of them. My goal: find out what kind of data might be useful to them.
Lesson #1: Think about what motivates you.
I knew that even if I didn't get into The Data School, if I could help someone out with a problem they have, I would feel excited and motivated to find a solution. I would be proud of my work regardless of the outcome of my application. Take time to think about what would make this process worthwhile to you, even if you don't get your ideal outcome.
Lesson #2: Don't be afraid to iterate, and ask different questions
The first time I asked my friend in the climate space what kind of data would be helpful for an upcoming project, he gave me a list of data he wished was gathered. Through my conversations, I learned that asking people about the kind of data they would like to see can often lead to requests for data that just doesn't exist, and that wouldn't work for my project!
Lesson # 3: Practice getting to know data quickly
Now, this is something I didn't do, that I would do now since hindsight is 50/50. When looking at a data set, try to describe 'What does one row represent?'
In this data set about dog licenses in NYC, each row represents a unique dog license. For each dog license, we can see information about the associated animal (name, breed, gender, birthday) and information about the license (issue date, expire date, and zip code).
Practice exploring data from the lens of 'what does each row represent in this data set?' Open up multiple data sets and see if you can answer that question for yourself. That will help you make sense of the data and determine if you might have fun playing with it.
Lesson #4: Don't be afraid to pivot
My original plan was to build a food security dashboard for a local organizer I know who runs a project rescuing food in NYC. After downloading a few datasets and realizing that the data was incomplete, or wasn't able to address the questions she had, I accidentally happened upon a data set about dog licenses in NYC. As someone who frequents a dog park multiple times a day, I knew this was the kind of data set I could have fun with.
Don't be afraid to download a data set, play with it for 30 minutes in Tableau and decide to go in a different direction.
Lesson #5 : Give yourself a deadline.
Our world is full of data, and even the free stuff is vast. So my advice to you in picking a dataset: give yourself a deadline and just pick one.
I ended up letting go of the idea that my dashboard might be useful to a user, and realized it's utility was to help me learn! Through building my dashboard I got to connect with my neighbors (because getting eyes on your project is essential), I got to learn about different chart types and visualization best practices, and begin learning a new tool (I had never used tableau before the Meet & Greet just 2 months before the application deadline). From that lens, it is impossible to pick a 'bad' data set. Even the Makeover Monday datasets might have as few as 20 rows, and 4 fields (columns).
Just pick one. It's not actually about the data set, it's about your process learning more about it. Besides, at The Data School you won't always have data you're excited or knowledgeable about, but your job will still be to make sense of it.
Good luck!
If you're thinking of applying to The Data School, just go for it. You'll learn a lot about yourself just through the application process. At the very least, you'll have started to build your portfolio on Tableau Public and receive valuable feedback, so if you're serious about this career path, building that up is important.