• No categories

Blog: The Challenges of Data Journalism : Interview with Jelle Kamsma from

Data Journalism


One thing we haven’t properly discussed yet, here at the Impakt blog, is data-driven journalism. If websites like WikiLeaks and open data initiatives all over the world are making information more and more available online, it doesn’t mean the regular reader or citizen can make sense of it. For this reason we need data journalists like Jelle Kamsma, who has been enriching and illustrating the articles at with specific infographics for a few months now. I met up with him for a lunch and a chat at the website’s headquarters, where he told me about what he’s doing and shared some interesting considerations on data journalism in general. Read on.


When you started working for, were there many other online magazines in the Netherlands with a data-driven journalism department?

Well, certainly not online magazines. I know that NRC had a data journalist working for them, but I think really was the first online publication to hire someone for that. Now it’s a bit different, because there is also Sargasso – which employed someone last week, and also works together with a news agency, ANP. It’s a really new territory, but it’s coming more and more now. And not just online publications, but also newspapers, or television broadcasters like RTLNieuws. But overall, in Holland, I think you can count them on one hand.

And it all started, say, within a year?

Yeah, I’m not sure when NRC started, but it’s a really new terrain. There is also no education for data journalism, and you need very specific skills, so everyone is kind of discovering it and trying out new stuff.

What was the hardest challenge when you started?

Well, there were a couple. First a technical one: you have a lot of different things you wanna do and a lot of technical skills to build up. Learning how to program, finding the right tools to use, and so on. That was the first challenge. Now that I feel I have some grip on that, that that is working, and I think the biggest challenge is to find actual stories in data. is really focused on news, not so much background stories or future forecasts, but hard news. Data journalism is usually very suitable for background stories, to put news into context – which I’m also doing a little at – but what they really expect from me is to find my own news and find new stories out of the data.

What kind of tools do you use?

My function is twofold. We get a lot of articles from press agencies, sometimes about statistics, and I try to find the relative data and make an appropriate visualization. I useFusion Charts for that, a paid software. For mapping geographic data I use Google Fusion Tables. It’s free and it gives you a lot of possibilities to map data. You can go on Google Maps and make polygons, paint them in different colors, import different shapes, work by province, city, even neighborhood or postal code. Lots of cool stuff.

Do you ever feel like these tools are not enough for the type of story you’re trying to tell?

Yeah. Of course I look up to the New York Times and they have a team with a developer, a couple of journalists, a designer… and they can make very cool applications. That’s not within my reach yet, but I can do most of the things I want to do. If I wanna use Flash, though, I have to keep in mind that it doesn’t work on iPhones and iPads, so they want me to use Javascript. Before I came people here also usedGoogle Motion Chart to follow the elections, how the polls were shifting and so on. I also use Impure, pretty cool stuff. It’s basically like a programming language, but it’s visual. You have all kinds of modules that you connect with each other in a physical environment: you can do all sorts of calculations on one hand and different visualizations come out on the other hand.

What response did you get in terms of readers, comments, and so on?

Pretty good. People are happy we do more than aggregating articles, that we try to make our own news. Also, if you make a mistake in a graph, you get immediate feedback on it. is very quick, you make something and you put it online, so we get a lot of feedback from readers.

Infographics are very easy to share via social networks these days and they’re all over Twitter and Facebook. Do yours get further exposure via social media in the Netherlands?

For Twitter we just have a counter in our articles, you can click that and you see all the tweets linking back. For blogs it’s harder, but they pick some things up. There are some that deal with data journalism, popping up also here in Holland, and they link to my articles. I was very proud that I made a visualization about the world population, using Impure, and when I checked their blog a couple months later I saw they referenced my article to show how I used their tool.

Do you think data-driven journalism has its limits? If yes, which ones?

Yes, first of all I don’t think we should look at it as an entirely new job. The skills I’m learning are becoming more and more important, but they will be integrated into the rest of the newsroom as time goes. I think data should be more part of the journalistic process than it is now, but the focus should always be on the stories. People are trying to disclose data and just put it online, but I think it’s really the journalist’s duty to find meaning in it. I do think you need to put the data online for people to check or look for further correlations, but the journalist has to find the story and write it down, even without a visualization.

Do you have any other developments coming up in your department at

The focus is to make your own news, we have a lot of contracts with press agencies, but we wanna find new people and make our own stories. For example I’m working with Lucas [Benschop], our political editor, for something about the government’s budget, which just leaked. You can apply data journalism to every field: politics, economics, culture… everything.


Thanks to Jelle Kamsma for his kind collaboration and interesting viewpoint!