Tyson Evans and Erica Futterman work on  newsroom strategy for The New York Times. The Times is pivoting from a centralized to more of an  embedded team model

Tyson Evans and Erica Futterman work on newsroom strategy for The New York Times. The Times is pivoting from a centralized to more of an embedded team model

Do you want to be ready for a startup in your newsroom, for new research methods, immersive storytelling, and product thinking? Then open doors for people with different skill sets. Invite coders, statisticians, bio-engineers, designers and {INSERT YOUR WILDCARD DISCIPLINE HERE} to work with you. But you’ll also need a fresh tool set and new management approaches.

Working with and within an interdisciplinary team at BR Data, the data journalism team of the German public broadcaster Bayerischer Rundfunk (ARD) for the last few years (yay, BR Data!), I came to Harvard last summer as a 2019 Nieman Fellow with a lot of questions about how to do it better. Here are some of the answers I found.

I divided my attention during my Nieman year among interdisciplinary newsroom management, automating the news, and algorithmic accountability reporting. I got to talk to a bunch of smart people who tried different approaches to insert interdisciplinary teams in their news organizations. I’m immensely thankful for their insights and the experiences they shared with me. You’ll find a list of all contributors at the end of this article.

Working with people with a different mindset and skillset from yours is one of the best experiences I’ve had in my career. And it can be a pain in the ass. I’m sharing this ambivalent feeling with most of the people I’ve talked to.

Because it’s a lot of trial and error. Because you depend on one another AND on the strategy of the news organization as a whole, which comes down to the nitty gritty details of your everyday work. Do you get server access, or do you have to ask the IT department every time you want to work on your story? Are you bound to a rigid content management system, or can you build your own storytelling framework to experiment with new formats? Do people with wildly different backgrounds really have to have the same journalistic credentials as every other person in the newsroom?

All this can cause a hell of a headache. But in the end it’s worth it. Together you’ll be making a much better product and most probably having more fun. That’s why this is a manifesto.

If you’re setting up an interdisciplinary team in a newsroom, these ideas, experiences, and tools will hopefully make your life a little easier:

  1. Define your mission.
  2. Grow your team organically.
  3. Fit a model to your newsroom.
  4. Hone a digital strategy and mindset.
  5. Work on a common language.
  6. Manage your skillsets.
  7. Live in beta mode.
  8. Develop your own tools (below are some you can steal).
  9. Engage in job development for news nerds.
  10. Experiment A LOT and learn from your failures.
    (I know, you know — but this is so important.)

1. Define your mission

“People in an interdisciplinary team need to fit together like puzzle pieces,” says Cameron Hickey, formerly team lead of the Information Disorder Lab, a project that ended in April 2019 and had been based at Harvard’s Shorenstein Center. His team worked with open source intelligence experts, developers, academic researchers, and reporters to identify misinformation on various social media channels during the 2018 U.S. midterm elections. “That’s why everything depends on a solid mission you’ve given yourself as a team.” In his experience, the line of dependency looks like this:


Having a well-defined and narrow mission helps to define the roles in a team with various disciplines. The roles translate into people, and only then you can plan on your day-to-day work as a team. “Our mission was too broad in the beginning. We also had too many people doing the same stuff,” Hickey recalls. So they adjusted their mission and tried to assign clearly defined roles per project.

The more focused your mission, the better it is. “Our goal is investigative data journalism” works out very well in terms of clarity. Everyone knows where the focus is. If you have more than one goal, try to formulate priorities, like “investigative projects beat non-investigative projects, interactive storytelling beats non-interactive storytelling.” It works as a guideline to decide which projects to take on and how to publish them.

Another thing to keep in mind: The mission will change over time. And that’s good, because down the road you’ll also get a better understanding of the problem you’re solving. That also allows you to hire more specialized people: “Starting off our project with generalists has proven to be good,” says Hickey. “The more you understand the scale and scope of the problem the more you should hire experts.”

2. Grow your team organically

2019 Nieman Fellow Francesca Panetta has put together a lot of interdisciplinary teams in digital immersive storytelling, most recently as team lead of Guardian’s VR studio . She usually had tight timelines: “You often have to launch these things fast, and you scramble to put a team together. But if you don’t know the specifics of your project, which often you don‘t right at the start, you won’t know what skills and therefore what team members you are going to need.”

It’s smart to grow an interdisciplinary team organically over time instead of launching it from 0 to 100 at once. Sometimes, you don’t have the choice —but if you can, lobby for more time. “If you have to build a team at once you will have a lot of turnover,” Hickey says. “It’s highly unlikely the team will work out as the mission and the team have to grow together.”

At BR Data, we started off very small and had the chance to grow with our mission,  which we changed a lot. We began as a Transmedia Storytelling Team and pivoted gradually to data journalism. During this journey we got better in understanding the direction in which we were heading and which skills were lacking. When we had the opportunity to fill a new position, we always tried to assess where we wanted to go next and fill the skill gap that felt most urgent. If we didn’t have the time to gradually build our team, we would have missed a lot of nuances and specializations that helped us building our DNA as the investigative data team we have today.

3. Fit a model to your newsroom

There are a few newsroom models out there that have been tested for interdisciplinary teams. But does the model matter at all?

To make it short: It’s essential. It determines how your team is embedded within the news organization, how people work together, who is setting the agenda, and to whom team members are reporting.

There are basically two poles between which most newsroom models shift:

CENTRALIZED  <— — — — — — — — — — — — — — — — — — — > EMBEDDED

CENTRALIZED = an independent team with permanent members

EMBEDDED = members of the team work embedded in other teams but regularly touch base to have an exchange among peers

Erica Futterman has seen a variety of digital newsrooms. She’s been at, among other news outlets, MTV News, BuzzFeed, and right now as deputy director, newsroom strategy operations at The New York Times. “By and large, everyone tries to solve the same issues, even if they have different models. The good news is, you can actually learn a lot looking left and right,” Futterman says.

Both models have upsides and downsides, according to Futterman: “If your team is centralized, you have fewer issues coordinating people and setting an agenda, as there usually is one team lead instead of different editors people are reporting to. It can be a challenge to keep things fresh, to pull in new ideas and not to isolate innovation happening in this team. If members of your team are embedded in other teams, it’s usually easier to keep things fresh. But everyone will have their own stakeholders and bosses to whom they have to report back. So here it’s important that everyone is on the same page right from the start, so that the day-to-day work is as frictionless as possible.”

+ The odds are good for an experienced, well-practiced team
– It’s a challenge to spread the innovation within the company

+ Keeps things fresh and spreads innovation
– Makes focus, coordination, and agenda setting hard

The reality is somewhere in-between these two poles and correlates very often with the size of a team and the depth of specialization. The more you target a certain specialization, the more you need a centralized team of experts to develop this skill set. The bigger a team, the more you can be flexible about embedding and centralizing — or trying both at the same time.

The New York Times has started to pivot from a purely centralized to a more embedded model by preserving large expert teams like Graphics, Interactive News, Digital Design, and Photography. At the same time, experts from those teams enrich permanently interdisciplinary desks assigned to beats like Climate, International, and Opinion. Hannah Fairfield came from the Graphics Department and now is leading the Climate Desk. “These teams are driving coverage instead of reacting,” says Tyson Evans, senior editor for strategy and product at the Times. And they are pushing new approaches: “Visual journalism and data journalism were the exception, not the norm. We’re trying to change that.”

The code-loving newsroom at ProPublica is more on the left side of the axis. They’re relying on their centralized expert teams, like the Data Team, led by Ryann Grochowski Jones, or the News App Team, led by Sisi Wei. For projects, they form non-permanent ad hoc teams with reporters from other parts of the newsroom or from other newsrooms they’re cooperating with for a story. Experts are hired by their team leads who also set the agenda together with their teams. “For skill growth and innovation in a field like news apps or data journalism, you need a strong connection to your group of peers,” says Wei. “It’s unbelievably hard to push innovation if you’re the only expert in a group.”

The Washington Post believes in ad hoc teams as well. Their newsroom is a matrix of topic-based vertical areas (politics, international affairs, economics, style, sports) and genre-based horizontal areas (design, photo, video). They use ad hoc teams to pull together people for non-permanent interdisciplinary teams that work on projects like virtual reality or automated storytelling. “That makes innovative dissemination happen more organically,” says Jeremy Gilbert, director of strategic initiatives. If those projects are successful they get a more permanent status. The Lily, a female millennial-focused publication, and the paper’s Snapchat team were born like this.

Before her 2018-2019 Nieman year, Kaeti Hinck led a visual storytelling team at The Washington Post and has tried both a centralized and an embedded approach. Neither extreme worked for her team, so she chose something in the middle: “We were happiest working as a self-standing team that brought others in.” That’s what I would claim for BR Data as well.

Wei stresses that there’s much more innovating to be done in terms of how can you maintain that expert community while giving people specific focuses in the newsroom. She’s perfectly right! So if you have an example for a model that’s useful for discussion, please let us know in the comments.

4. Hone a digital strategy and mindset

For Aron Pilhofer, the key to making an interdisciplinary team fly is: Digital must be core to a news company. “It’s more a digital strategy and mindset that results in the right decisions and structures than the structures themselves.”

Pilhofer has led digital strategies for major news organizations like The Guardian and The New York Times and is now professor in journalism innovation at Temple University. Over the years, he has developed a kind of a barometer that indicates how friendly an environment at a news company is for interdisciplinary teams to push innovation: “How far along is the company in shifting the values from the traditional to the digital product? Is the digital product at least valued as much as the non-digital product?”

If the answers are yes, interdisciplinary work and people with nonbinary skill sets have a good basis, according to Pilhofer.“But if different skill sets are not valued, it’s extremely dangerous to organize a distributed team,” says Pilhofer. “Then it’s better to create a protected environment.”

This opinion is shared by the overwhelming majority of the people I spoke to. Nearly everyone has had this experience: a great team limited by a non-digital environment.

One of my teammates asked, “Yeah, but what does a protected environment look like?” Well, it depends on the newsroom. It’s a huge advantage to have everyone in your team on the same payroll. When people report to different bosses, it’s hard to keep things together and find a common agenda. Also, try to bridge differences. Talk to other departments to find out what drives them and try to really cooperate. Our aim should be to solve some of the others’ pain points . In my opinion, the only way forward for a siloed team is to show how you can be useful for the whole company.

For Scott Klein, deputy managing editor of ProPublica, it’s obvious that each team member is an equal part of the newsroom: “Everyone in the team is a journalist with different skills.” He was able to build the ProPublica newsroom from scratch and implement this attitude from day one. But he is surprised how long it takes to change the minds in most legacy newsrooms: “I think we need to have a much broader conception of what journalism is.”

5. Work on a common language

“It took us months to understand what everybody did,” says Shaul Amsterdamski, who led a journalistic motion graphics team for Israel’s Public Broadcasting Corporation before his Nieman year. If he could do it again, he’d organize a bootcamp at the beginning to outline how the team works.

Ana Serrano calls that “sharing domains.” Serrano is chief digital officer of the Canadian Film Centre and managing director of ideaBOOST, the Centre’s digital accelerator: “The ability to understand the skills of the other people and to translate your own language into theirs is key.”

For Serrano, sharing domains means not only talking about your skills, workflows, and your professional background, but also understanding how people process information. When she builds a new team or adds new people, they spend at least two full days in show and tells — especially if time is scarce.

“It’s a huge mistake to dive into a project without taking the time. In the end it will definitely save you more than it costs you.” If she has to be quick, she skips the fun team-building exercises: “We’ve spent 10 years experimenting with ice-breaking methods. And it really is not as important as sharing your domains with the team.”

And yeah, it’s work, and not only when you start off as a team. You constantly have to share knowledge and integrate that into your day-to-day work as a team. (See tools under point 8.)

6. Manage your skill sets

The less translation work necessary, the more frictionless the workflow.

That doesn’t mean that everyone in an interdisciplinary team has to have the whole skill set. That would only lead toward generalized knowledge and make specialization more difficult — and, in the end, teams less diverse.

But you have to make sure that there are overlapping skill sets among the different crafts. “Teams fail when different tasks don’t overlap,” says Scott Klein of ProPublica. “Let’s say, well, I’m going to do the data and you’re going to do the interviews and the writing. And we’ll write memos to each other or maybe even not that. That’s how mistakes and incongruities are born. It’s important to really collaborate.”

If you take a closer look at the methodology of BR Data’s and Der Spiegel’s investigation about discrimination on the German rental market you can feel how intertwined the different experts of an interdisciplinary team work. From the first idea to the final text production, investigative work, data analysis, storytelling, and graphics went hand in hand. Programmers and researchers worked together on the leading questions that took us to the data gathering. After the data analysis, everyone worked on versions of the text and in parallel on the graphics. A ton of Slack communication, Skyping, and real-life meetings between BR Data in Munich and Der Spiegel in Hamburg made it possible to tell this story about a society excluding people with foreign-sounding names from the rental market and denying equal access to housing.

For that kind of collaboration, everyone in a team has to share some common ground with other team members. How can you do that? There are already people in journalism who refer to themselves as nonbinary journalists: journo-coders, interactive designers, you name it.

Also teams can build structures to share knowledge: Coders giving Github intros, writers teaching investigative methods, designers and writers developing a design language together.

But you have to be honest about this need, and you seriously have to make time for this kind of exchange. In the end that will — again — save you more time than it costs.

7. Live in beta mode

If you’re a perfectly agile project manager, skip this point.

Work in versions. We’ve all learned this because most interdisciplinary teams live in a fast-changing innovation environment. But it’s kind of hard. I have to admit that even after years as part of an interdisciplinary team I have to remind myself VERY often to actually practice it. It’s useful for all the parts of interdisciplinary teamwork: project plans, research hypothesis, workflows, team structure — and even the mission for the team (See point 1).

Voilà the ideal workflow: prototyping, testing, revising — repeat.

It means trying out ideas by breaking them up into small portions of work, taking first steps, evaluating the result, and deciding as a team to go on or dismiss the idea. Key is responding to change. Coders build beta versions to test software, and we’re borrowing that idea from agile developing and methods like scrum to stay flexible in projects.

But how do you integrate that mindset into your newsroom workflow?

Mark your schedule to re-do things and develop a habit to go over as a team the current projects (See tools in point 8).

Also, involve externals in testing your prototypes to discover your blind spots. Magnus Bjerg, who as a 2019 Knight Science Journalism Fellow at MIT also did research on interdisciplinary teamwork, remembers several projects for Danish TV2 that were turned around after testing them with users. This story about an extreme running talent in Denmark got a completely new beginning after letting colleagues from other departments scroll through it: “We thought everyone was hooked from the beginning, but it was obvious that they just finished it because we asked them to.”

He also stresses how important user feedback is:  “When we created a candidate test for our last local election, we added a link to a simple feedback form at the bottom. We quickly learned that a lot of people were angry that we didn’t show all the candidates in our test. The only reason we didn’t show that was because they hadn’t answered our survey. So we quickly added those candidates as well with a ‘hasn’t responded yet’ label to clearly communicate that we wanted to show their opinions also.”

The more experimental you work, the more feedback you need, says Magnus: “When working in an interdisciplinary team you often push the envelope for what is possible. That in itself creates a need for even more feedback and user testing because you are in essence doing new and unproven things. It will most likely be a bit off target the first few times.”

Also, you yourself are your best beta version. Wei tries to keep herself and her team at ProPublica in an ongoing refreshment loop by diving into projects, trying to figure out problems, learning by solving them: “Let people teach themselves and let them know that they don’t need to have the answer to be able to pursue a problem.”

8. Develop your own tools

Each team is special in its own way, and it works best to develop your own workflows and tools. But here are some that recurred over and over again during my interviews. They seem to work for a lot of people — feel free to choose and pick.

All these tools refer to Hinck’s key value: “Do the work, don’t talk about the work. Avoid unnecessary meetings as much as possible.”

And sometimes meetings can help you avoid more meetings 🙂


= meeting no longer than 10 minutes each morning

  • serves as an update: everyone says what they’re up to today
  • no discussions
  • if discussions are necessary, they’re assigned to the people involved
  • gives visibility to what people are doing
  • puts everyone on the team on the same page
  • saves time for project managers
  • borrowed from the agile method scrum


= meeting no longer than 1 hour once a week

  • establishes the habit to work in versions and to question your methods
  • talk over each project you’re working on
  • focus on the changes since the last meeting
  • discuss briefly, if necessary
  • reschedule longer discussions

Project Manifesto

= central document for each project

  • outlines the purpose of the project
  • keeps you focused on your goals
  • for bigger projects, also outline the phases of a project
  • integrate the personas of the users you’re targeting

Impact Sheet

= document for desired and accomplished goals for each project

  • outlines the impact you wish to accomplish for a project
  • helps to stay focused on your goals and to assess your impact
  • helps to understand where and why things have pivoted
    (which can be a good thing)
  • fill in data before and after a project
  • use it for feedback after each project


= project reviews

  • evaluating what went well on a project and what didn’t
  • analyze why things went wrong
  • helps a team to evolve and grow
  • you learn best if you have room to understand your failures


= project overview

  • digital or physical version
  • visual overview of what the team is working on
  • can be filtered by persons and projects
  • can serve as an inclusive tool for the whole newsroom

For team communication, Slack is the best way to prevent everyone from drowning in emails. If you’re collecting and sharing information, some teams also use tools like Microsoft’s OneNote, as it allows more complex indexing.

A team space can be an asset. Team members sitting close to each other make additional meetings unnecessary. But you have to establish quiet time. Perfection is to separate quiet and noisy — but only a few newsrooms are lucky enough to have that. I saw a pretty good example in the new Boston Globe newsroom.

9. Engage in job development for news nerds

There’s an emerging discussion about the lack of career opportunities for news nerds. This discussion at the Journalism Festival in Perugia touches the issue in interesting ways.

Integrating people with diverse backgrounds doesn’t end by giving them a seat at the table. If you’re passionate about your job, you very likely want to have a perspective on what to do next. But senior editor or leadership roles in most legacy companies are reserved for people with a traditional journalistic background.

Even industry leaders struggle to make space for new roles. There are exceptions like Fairfield at The New York Times, and Amanda Cox, the data editor of the Times’s Upshot . But it’s always the same examples that are quoted, which shows how rare they are.

Erica Futterman from the Times points out that career planning in the news industry in general is not easy: “So many jobs exist today that didn’t five years ago. That’s both exciting and challenging.” But people with a nontraditional journalistic background have to be especially pro-active: “There’s no predecessor to model their career after,” says Futterman. “Desk heads are mostly still people who come in with more traditional reporting jobs. That’s definitely something we’re wrestling with.”

So what would help? Certainly keeping people with diverse backgrounds in the loop and asking them what they need to thrive in the newsroom. Involving them in developing training for the next generation of journalists. And integrating people in Human Resources who understand background diversity as a must in newsrooms.

The good news is: If you succeed, it’s a big Unique Selling Point for your newsroom that will attract talent. And that’s crucial. Because, for now, there aren’t tons of people with a nonjournalistic background pushing into newsrooms. It’s a pipeline problem: People who fit in those roles don’t even know that there might be a career in journalism for them. And only a few universities, like Columbia Journalism School’s Lede Program, specialize in training journalists who might fit in those roles.

Large newsrooms like The New York Times have recognized this scarcity and are pro-actively looking for talent, says Futterman: “The Times has been focusing its hiring efforts over the last few years on journalists who come from more nontraditional and digital-first backgrounds.”

But as long as people are still hitting a glass ceiling for news nerds, there will never be enough talent. So let’s change that!

10. Experiment a lot and learn from your failures

I didn’t want to let you go without this: Interdisciplinary teams need room to experiment and a management that is serious about giving them space to fail. Of course, you have to learn from your failures, as The Washington Post’s director for strategic initiatives Jeremy Gilbert puts it: “We try very hard to reward not only success but clever attempts. So that we can be wrong —  but that can be valuable, too.”

Thanks to all the smart people who talked to me and gave me insights in their experiences, workflows and opinions! The material was gathered in interviews and in a seminar I organized at the Nieman Foundation with my fellow fellows.

  • Shaul Amsterdamski, 2019 Nieman Fellow and most recently lead of a journalistic motion graphics team for Israel’s Public Broadcasting Corporation
  • Magnus Bjerg, Fellow at MIT and digital projects manager at the interdisciplinary news team at TV2 Denmark
  • Matt Carroll, professor of journalism at Northeastern University
  • Tyson Evans, senior editor for strategy and product at The New York Times
  • Erica Futterman, newsroom strategy editor at The New York Times
  • Jeremy Gilbert, director of strategic initiatives at The Washington Post
  • Ryann Grochowsky Jones, deputy editor data at ProPublica
  • Cameron Hickey, former team lead of the Information Disorder Lab at Harvard’s Shorenstein Center on Media, Politics, and Public Policy. The lab ended in April 2019. First Draft continues the work in New York and Hickey remains at Shorenstein researching misinformation.
  • Kaeti Hinck, 2019 Nieman Fellow and most recently an editor at The Washington Post
  • Scott Klein, deputy managing editor of ProPublica
  • Francesca Panetta, 2019 Nieman Fellow and most recently team lead of The Guardian’s VR lab
  • Aron Pilhofer, professor of journalism innovation at Temple University
  • Ana Serrano, chief digital officer of the Canadian Film Centre and managing director of ideaBOOST, the Centre’s digital accelerator
  • Sisi Wei, assistant managing editor at ProPublica and lead of the News App Team

Please leave feedback in the comments or at Twitter — looking forward to a discussion!

Further Reading

Show comments / Leave a comment