The story of a transformation, in three years


How Julius Troeger, a journalist working for Berliner Morgenpost, managed to create a string of impressive data interactives and won the support of his publisher for a small, but growing interactive team in the newsroom.

If data journalists could write a wish list, it would look like this: to have the time to work on interactive stories - days, even weeks, not just hours, to be supported by a skilled and capable team, encouragement from upper management, and, finally, a budget large enough to achieve all of the above.

foto-julius-troeger-klein.pngJulius Troeger, a German journalist working for Berliner Morgenpost, can tick off all of these items. He is heading a small, but growing team, which has already published a great string of award-winning stories.

It took him roughly three years.

Today, Berliner Morgenpost is known for its outstanding data interactives. Despite being a local, regional publication, the quality of its data journalism is widely recognised - even journalists from competing larger publications like Der Spiegel acknowledge the quality of its work.

Creative output, strong usage numbers, and a growing expertise in data journalism enticed Funke Mediengruppe - one of Germany’s larger publishers - to acquire the title in mid-2014.

Importantly, the new management understood the value of the work done by the interactive team. In a press release from November this year, the group announced that starting in 2016 the interactive team headed by Julius Tröger will be not only be extended, but will also support other newsrooms in the group. 


Julius, you are one of the few data journalists working for a daily newspaper in Germany. Can you briefly describe your journey so far?

After I was a trainee at a local print newspaper, I worked there as a digital journalist focusing on web videos. My job at the Berliner Morgenpost began in 2010. At the same time I began my part-time post-graduate studies in computer science. After interning in 2012 at ProPublica and The Guardian US in New York I started my own interactive team at the Berliner Morgenpost. Since 2014, developers, designers and other journalists work together in the newsroom. We regularly publish interactive applications, graphics and maps and experiment with new journalistic formats. Next year our team will grow and we are going to publish interactive in cooperation with other newsrooms by our publisher, the Funke Mediengruppe in Germany.

Why did you get into data journalism in the first place? What is your motivation?

Actually, I always wanted to be a programmer. In elementary school I wrote small programs with Turbo Pascal. Later, however, it was more words than code that have inspired me. I then decided to go for journalism. Today I have the great fortune to bring together both. After I visited a data journalism conference in 2011 I was so inspired that, together with a colleague, we started to work on our own projects right away. It excites me, when a scraper runs to get information or when our readers share their personalized news from our interactives. Data journalism techniques helped us getting several exclusive stories.

You became "Journalist of the year" in 2014 - did that change anything for you?

I think it drew a little more attention to the advantages of working with programmers among traditional journalists. In the same year our team was also on the cover of the German media magazine “Medium Magazin” with the headline “Nerds in Newsrooms”. After that other newsrooms sent colleagues to intern in our interactive team to start their own data journalism unit.

You and your team have by now published quite a few interactive data pieces - which one is your personal favorite?

My favorite project is “M29 – the bus route of contrasts”. The bus route cuts straight through Berlin. It starts in the villa districts in the west of the City, passes through the inner city areas, and ends in Berlin’s trendiest districts. We collected and processed data regarding the neighbourhood and local residents for every stop along the bus route. With the interactive application we highlight the social differences of Berliners.

Which project looked easy and became complicated later? Or the other way round...

We just launched an interactive data visualization project in 3D. It shows how Berlin has developed since 1990 - and how it will look in the future. We had to deal with such a large amount of data and file types that I`ve never heard before. It took us a lot of time to figure out, how deal with that.


When you think about a new project,how do you balance data vs. storytelling?

I'm working with a great team, with which I can think each and every story format from scratch. In my opinion we should always surprise our readers, either with a fascinating topic or story such as our finding, that there are flights without passengers from one airport to another in Berlin or with an experimental format such as our single page about the delayed Berlin airport where we translated the timetable’s complicated technical jargon into a comprehensible construction checklist.

Which technologies do you use -  visualization libraries, maps, etc.?

We have basic reusable code what we call our “Starterkit” for articles and maps. For maps we mainly use Leaflet and Mapbox. We also had several projects now with turf.js. For charts we mostly use D3. We use QGIS and GDAL a lot for geo data. Excel and R are our tools to clean and analyze data sets. For bigger projects we use MongoDB, for smaller ones NeDB. Our code is mainly written in Javascript and NodeJS.

Given the availability of data in Germany, are you more often angry about complexity of getting information or is that improving?

As local journalists in Berlin we have the privilege to work in a city that today provides more data compared to other parts of Germany. Five years ago we had to draw districts on maps by hand. Now, you get shape files or machine readable data formats from authorities more often. Also there now is a better communication between the statisticians or data experts working for the city administration or the statistical office.

How big is the team you work with?

Next year our team is growing. Then we will be two full-time journalists, two part-time developers and one designer. There are always other colleagues joining us occasionally

If you would give advice to another journalist interested in starting in data-driven journalism? What should she/he focus on?

Everything we do is possible only because programmers share their code and journalists and designers share their experience with others. So just start reading blog posts and tutorials. With that knowledge try to work on your own project. If you get stuck, there is always someone on Stackoverflow or the NICAR mailing list that helps you.

There is a big debate about whether journalists should learn to code. In your opinion: What is achievable? How far can one get with coding in the newsroom?

I think as a data journalist you should learn just enough to code, so that you can obtain information and clean data on your own in your research process. That includes writing small scrapers, working with R or knowing how to automate recurrent tasks. But you should always hire developers to that you can focus more on the journalistic part than the technical part of the story.

Some examples of data stories published by Berliner Morgenpost

A more extensive list with many more examples can be found here at Julius Tröger's website or directly on the website of Berliner Morgenpost.