Data journalism might not seem like an obvious coding job, but if you have data science skills and a passion for storytelling it could be the perfect niche. News organizations and media outlets are looking for people who can look past raw numbers to find a compelling story and visualize it.
We spoke to Will Coulman, who at the time of writing was a Data Journalist at the business and tech news outlet Sherwood Media, to learn more about what Data Journalism is and what type of skills you need to get a foot in the door.
Learn something new for free
What does a Data Journalist do?
Simply put, Data Journalists “try to visualize the news,” Will says. A big part of being a Data Journalist is presenting dense, structured information on topics (think inflation, finances, and politics) in an approachable, engaging way. Visualizing data through charts, graphs, animations, or diagrams makes it easier for news audiences to conceptualize the information. Plus, visuals often make the findings more interesting and accessible than raw numbers alone.
For example, Will’s team might use Sankey diagrams to report on a company’s annual revenue and profit margin, showing how a company’s revenue comes in one side and then the profit and cost come out the other side. Take a look at the diagram in Will’s recent story that came out of an EA Games earning call. “It’s a simple chart but it captures the whole business in one image,” he says.
Reading and exploring examples of data journalism is a great way to wrap your head around the field. In addition to Sherwood, you can check out 538 for political data journalism, or this digest of outstanding examples from around the web. While being a strong writer is obviously key (this is a journalism role after all), the story is built on the data, so having a strong foundation in data analysis is vital too.
What does a typical workday look like for a Data Journalist?
At the beginning of a project, Will’s team will collaborate to develop a story idea. They’re looking for a few key things: the idea must be topical, interesting, and have relevant data to support it. “We will pick a topic like OpenAI’s revenue or US inflation, grab the raw data, and then really mess around with it and have fun,” says Will. That could mean using pivot tables to slice the data multiple ways to see if interesting patterns or outliers emerge. Understanding the data, assessing its quality and limitations, and identifying trends are crucial first steps. Data Journalists simultaneously have to determine what newsworthy story lies within the data.
Once a Data Journalist decides on a good angle for their story and gets the go-ahead from their editor, they might do further analysis, with some data journalists using SQL, Python, or Jupyter Notebooks. After analyzing the data, they create a chart to visualize the key data. “We try to keep it simplistic as well — we won’t want to overwhelm readers with too much information in the chart,” says Will.
From there, they usually write a brief story (250-300 words) to accompany the data visualizations. That might mean conducting further research or at some organizations, conducting interviews to round out the story, which provides more context for readers and gives the piece credibility. Depending on the news outlet or scope of the story, a Data Journalist might partner with another reporter who handles the bulk of the writing process.
There’s a fairly even split between data analysis and writing, “but it really depends on how complicated the data is,” Will says. There can be a lot of trial and error as you try different styles of data visualization to best tell the story. “By the time you start writing the accompanying words, you already have an idea of the flow of the story and aren’t likely to go back and start over like you might at the data visualization stage,” he says.
Do you need to be able to code to be a Data Journalist?
You definitely need technical skills to be a Data Journalist, but there’s plenty of great data journalism that can be done with good old spreadsheets. “We use Excel for most of our work,” says Will. “It does the job — there’s a reason why Excel is still a huge business for Microsoft.” Spreadsheets are extremely powerful software that you can use to sort and analyze data without coding yourself.
Like lots of technical roles, learning on-the-job is common in data journalism. In some organizations, you might be working with data stored in a SQL database, in which case some knowledge of SQL will be required (and there are also benefits to learning SQL even if your role doesn’t technically require it).
Python can be handy when it comes to doing more advanced data analysis and manipulation. Python’s beginner-friendly data science libraries and tools like Pandas, NumPy, Jupyter Notebooks, and BeautifulSoup make it easy to scrape data from websites and start analyzing it. You can also build charts with Python, Seaborn, and Matplotlib — our beginner-friendly path Visualize Data with Python will walk you through all the coding you need to know to start making data visualizations.
What other skills and qualities are important for data journalists?
Patience
Combing through data to look for trends can be painstaking work requiring a lot of attention to detail. You must be willing to dig deeper than the surface level to understand any potential faults or inconsistencies. Working with data can mean going down a rabbit hole only to realize there isn’t a compelling story to tell. “Oftentimes we spend hours on something and just end up chucking it out,” Will says, so patience is necessary to avoid getting frustrated.
Collaboration
Storytelling through data is a creative pursuit involving a lot of collaboration. Every article begins with a group brainstorm to develop the idea. “It’s always good to bounce ideas off of people,” says Will. “You need feedback to understand, ‘I see something this way, but does everyone else see it that way?’” Journalists frequently interview several people for a story to include diverse perspectives, so you need strong interpersonal skills to conduct those conversations.
A growth mindset
Learning on the job and staying up to date on new tools, techniques, and developments comes with the role, says Will. “I might be trying to analyze some data and struggling to wrap my head around it, so I’ll have to figure out how to do it in the right way while I’m working on the project.” Having a growth mindset and being endlessly curious can help you approach challenging tasks and keep building your skills.
How to get started in data journalism
There are a couple obvious ways into data journalism: adding data analysis skills to a background in journalism (we have a host of courses to round out your data skills), or building on a data science background by sharpening your writing and communication skills.
Whether you’re new or changing careers, many skills can transfer to data journalism, even without a tech or journalism background. Will has a Masters in Behavioral Economics, but got his foot in the door by applying to a small startup publishing a data storytelling newsletter. He’s since moved into an in-house data role on the editorial team at Knight Frank. His advice is to play around and produce your own content first.
“If you can find an opportunity, maybe at a small company, you don’t need an obvious background match, just a passion,” says Will. Learning to create charts and sharing them on social media will help you get feedback and learn how to improve while building your reputation. Even if you don’t build a large following, you will have a portfolio that you can share when applying for data journalism jobs.
Ready to try your hand at your own charts and diagrams? We have Data Science courses to build out your data analysis skills, including how to do data visualization with Python.