This article was published on gijn.org by Nils Mulvad & Helena Bengtsson in 2015 and updated in 2023 by our team
1. Begin with Small Projects – and Excel
When it comes to data journalism, it’s important to start small and use tools you are comfortable with. Excel remains a vital tool in data journalism, as it allows you to import and export data, clean, sort, and structure it before importing it into other tools. So, it’s essential to get comfortable with the basics before moving on to more advanced tools.
2. Learn the Tools of Journalists
As a data journalist, it’s important to understand the principles of calculating tools. If possible, it’s best to learn from a journalist who is experienced in this field, as they can help you relate specific functions to journalistic tasks.
3. Check, and Check and Check for Errors
Always assume that there will be errors in the data, even if you obtain it from a reliable source. It’s your responsibility to eliminate errors before presenting the data to the audience. Check totals, ensure that everything is included, and if you find any errors, go back and fix them.
4. Make Public Documentation of your Data-Work
When working with data, it’s important to document every step you take so others can follow your process and check your work. Sharing your data work with sources before publishing is also recommended as it helps in checking errors, agreeing on methods, and avoiding errors and criticisms of methods, which could take away from the discussion on content.
5. Use Errors to Get Better Internal Sources
When you discover errors in the data, use them as an opportunity to build better relationships with the people who work with data at the institutions or authorities that provided the data.
6. Get the Data
There are various ways to obtain data, including asking for it, downloading it from the web, scraping it, or making a Freedom Of Information request. Whatever method you choose, it’s important to develop and refine these methods just as you would with interviews, as they are journalistic methods.
7. Analyse the Data – Go for Stories
Data journalism should focus on telling stories. Analyse the data, find the stories within it, and tell them one by one. Use interactive graphics to break down each story into a clear and easy-to-understand format. Avoid mixing multiple angles and topics into long reads that could confuse the audience.
8. What Are Really the Story – Using Extremes
To find stories within the data, think deeply and check them against reality. The best way to describe stories is by using extremes. Use ranking as a tool to find errors and explain reality.
9. Data Journalism Sometimes Starts With the Story – and Sometimes with the Data
There is no right or wrong way to approach data journalism. Sometimes you start with a story and find data to support it, while other times you dive into the data to discover potential stories and then verify their accuracy.
10. Make Data-Cleaning Yourself
When there is no data available, structuring your own dataset might be the only way to create an incredible story. Don’t shy away from the hard work of data cleaning, as it gives you a better understanding of the data and potential stories hidden within it.
11. Get Rid of the Numbers
Data journalism should always focus on the human stories behind the data. If the numbers are overwhelming, focus on finding the humans who can best exemplify the data. Sometimes, zero numbers are the best way to tell a data journalism story. You can include the data in a graphic or keep it for later. If you can’t find humans to explain your findings, then they might be wrong.
12. Work Together and Share
The data journalism community has a long tradition of sharing tools and methods. This is essential to keep up with the speed of development in this field. Sharing is more common in data journalism than in any other area of journalism.
If you’re a business and would like to upload a media release, click here.