Twelve Lessons for Creating International, Data-Driven News Collaborations

This post was created automatically via an RSS feed and was originally published at


A new paper by William E. Buzenberg, Joan Shorenstein Fellow (Spring 2015) and former Executive Director of the Center for Public Integrity, explores the need for and impact of international collaborations between news outlets.

Buzenberg argues that although our world — and its resulting news stories — have become increasingly more globalized in nature, from finance to the environment to crime, most news outlets still find their scope restricted by nation-state borders and thinly-spread foreign correspondents.

Buzenberg’s paper details the founding, successes, and tactics of the International Consortium of Investigative Journalists, an organization that leverages big data and the skills of journalists in 65 countries to produce investigative reports.

The paper also provides practical takeaways for news outlets hoping to replicate such a model, excerpted below. Read the full paper.

Lessons Learned

Someday it may be commonplace for dozens or even hundreds of journalists and multiple news organizations to work together, sharing documents or massive datasets, in order to produce a series of investigative reports. Today, it is exceedingly rare. But the high impact of these mega-collaborations for relatively low cost means there will likely be others.

“Collaborations are the new reality, whether we like it or not,” ProPublica’s Jeff Gerth has pointed out. “Like most hybrids, they have costs and benefits.”

Media thinker Clay Shirky has written in his book Here Comes Everybody that the litmus test for collaborative productions is simple: “No one person can take credit for what gets created, but the project could not come into being without the participation of many.”

“Collaborations are the new reality, whether we like it or not,” ProPublica’s Jeff Gerth has pointed out. “Like most hybrids, they have costs and benefits.”

Indeed, the benefits, especially the impact, can be huge, but so can the difficulties in coordination with so many participants. ProPublica has partnered with some 120 different news organizations. But its president, Richard Tofel, says one reason Pro Publica prefers to partner with just one other news organization at a time, or possibly two, is because in larger collaborations “the transaction costs are simply too high.” Journalists, he suggested, tend not to want to spend their time “herding cats.”

The following checklist of essential requirements for collaboration is not aimed at small partnerships where one, a few or even several news organizations or journalists work together. There are many lists for what makes a good collaboration on that scale. Nor is this a list for distribution deals, where dozens or hundreds of news organization publish the original work of one or two.

This checklist is for the very largest collaborations, where a hundred or more journalists work together. Most likely their association will be virtual and online, whether around the country or the world. This checklist is derived from a number of documented reporter and editor experiences as well as the work done by the International Consortium of Investigative Journalists, or ICIJ. For its Swiss Leaks project, first published in February 2015, ICIJ made use of the largest collaboration on record: 170 journalists at 65 news organizations in 56 countries.

Here are the key lessons learned and some of the operational requirements for this scale of collaboration to work effectively:

  1. All collaborations need an editorial core
    Central leadership is essential. Central decision making means setting the agenda and serving as a communication hub, both necessary ingredients that can only come from a single center; this cannot be shared among many. Collaborations have fallen apart when there are two or three or more leadership nodes. Successful collaborations, such as the leadership from the ICIJ hub, are neither anarchies nor democracies but something more akin to benign dictatorships. They are not there to rule but to organize, instruct, coordinate, and be responsible for clean and shareable data and for timing of publication. An editorial core setting up systems, fielding questions, dispensing advice, offering assurance, maintaining calm, as well as a myriad of other operational, legal and strategic attributes, is critical for the work to flow.
  2. It helps if there are datasets or documents to share
    Overall, as ICIJ has learned, a good topic or subject is fundamental for a good collaboration. But one reason dozens of journalists might want to come together is often to gain something that they could not otherwise obtain. This includes combining forces to share a leaked dataset, or thousands of secret documents, or just being part of a global scoop. Creating a giant dataset from scratch, or building one from government data, can also drive collaboration. The data needs to be centrally cleaned and fact-checked and put into a format that makes them truly shareable and useful for multiple reporters in multiple locations. But whatever the source of the data or documents, these must be part of a significant and meaty topic. With an important theme, shareable and possibly secret information, reporters can more easily bond with the idea of a mega-collaboration and their part in it.
  3. Collaborations work better with two platforms
    New open source technologies make the practical orchestration of mega-collaborations possible if not easy. As the ICIJ team discovered, they needed two platforms to be effective, preferably encrypted: 1) for seeing, understanding and downloading the data as required for all participants, and 2) a platform for internal communications between the editorial core and the participants, and, even more important, among all participants. One of the great strengths of mega-collaborations is the exponentially greater opportunity to communicate across the network. Suggestions, questions, interviews, roadblocks, available video and breakthroughs can all be shared. This was a critical element in ICIJ’s Swiss Leaks project. The size of the network can therefore be a big plus.
  4. Money is best kept out of the equation
    Collaborations are cost effective. It is far easier when most members bear their own costs and their own liabilities for their work. Yes, the central editorial core needs sufficient funds to hire its own small team, handle the data and the central website, and cover the costs of the software, web servers, overall legal expenses and the like. But each collaborating reporter will need his or her own financial backing, most likely from the participating news organizations, which must buy into the project at the start. Some reporters may be under contract or working for a freelance fee in successful collaborations, as often happens in ICIJ projects. Hiring, say 50, independent reporters and paying each of them is a different matter, in essence work for hire that is cumbersome, difficult and costly. This approach also involves a centralized editing function and thus has far higher “transaction costs.” Conversely, if money is not a big factor, the focus of the collaboration is almost all editorial, along with timing and technical requirements.
  5. Have at least one face-to-face gathering
    It may be counterintuitive in our virtual and highly connected world, but collaborations are a “high-touch enterprise” and require face-to-face meetings. Such meetings still mean a great deal, again, as ICIJ has learned. Just as an online dating service can work well up to a point, an in-person meeting is eventually necessary and will tell a lot very quickly. For trust to start building within a major collaboration, it’s essential to size up the participating people, to see how they handle themselves and how they address the important issues at stake. Are they serious professionals and do you have a good feel for their strengths and weaknesses? Or is a reporter getting into something where the standards of journalism are not where they need to be and therefore the resulting project is not likely to measure up? There is no substitute (not even Skype) for a physical gathering where participants can look directly into the eyes of colleagues with whom he or she will be working, perhaps over many months. A core gathering can lay out the investigation and the evidence at hand, discuss the challenges and the payoffs, and perhaps provide training and expert analysis. Such a meeting will save time down the road when the work is hard and seemingly never ending.
  6. Some loss of central control is inevitable
    It would be extremely difficult for the editorial core of any mega-collaboration to be 100 percent responsible for every report from every collaborator. ICIJ learned this lesson after years of trying to edit and fact-check all ICIJ member material. However, the data itself must be centrally organized and fact-checked. The participating members need to be vetted and vouched for, making sure there is an acceptable level of quality within the group. But each news organization has to be responsible for its own reporting, editing, fact-checking and libel review in order for the collaboration to proceed on a timely basis and ultimately to succeed. This gives each participant the freedom and responsibility to do his or her own best work, based on the shared data or documents. But this freedom also means a necessary loss of tight central control, which makes some news organizations uncomfortable. They do not want to be associated with others over whom they have no editorial sway. This raises the critical issue of trust (see number seven below).
  7. Trust is learned; shared values are critical
    As Julia Stein, an ICIJ member and public broadcaster in Hamburg, Germany, said: “It was very unusual in the beginning to share everything with other journalists and not keep secrets anymore. Collaboration is a question of confidence and trust in others.” Trust is best built over time. The beauty of ICIJ’s series of investigations of secret offshore dealings is that each project built on the previous one, often drawing in the same reporters and news organizations. By the end, many of the reporters had been working together for three years and were comfortable with ICIJ’s methods and the skill level of most team members. A sharing of basic journalistic values is fundamental. Although mostly accustomed to working on their own, reporters can learn to depend on others, to trust people in the collaboration and those leading it, even if there is some variation in the practice of journalism in the various participating countries.
  8. Set aside competitive instincts and egos
    Joining forces and trusting other journalists does not come naturally to independent-minded “lone wolf” investigative reporters. Harder still, sharing secret information and nuggets of fact gleaned by one’s own hard labors is not a normal practice. It is only through experience and repetition that collaborating journalists learn to see the success of the whole is as important as their own. Again and again, when one publication sees the impact of 50 or more news organizations publishing at the same time, then they tend to see that the group can have much greater impact — and that they are a vital part of something bigger than themselves that can redound to the benefit of all.
  9. The group can help
    Working in a complex team configuration over many months tends to help reporters see the value of the group. Having a closed communication platform just for members of the group (encrypted in the case of ICIJ), fosters insider knowledge, valuable information sharing, and trading of video and transcripts of interviews and the like. Turning to the group for help often produces better results more quickly than any other method. As noted earlier, the power of a network itself is tremendous. The Swiss Leaks collaboration, with some 170 reporters, had the potential for nearly 29,000 communication combinations in the unlikely event that everyone made use of every connection. But using even a fraction of that network power proved itself many times during the complex investigation.
  10. Scale begets power
    Joint publishing in dozens of countries around the world can expand the impact of the reporting in any one country. As in the case of Swiss Leaks, the HSBC bank’s reaction changed in direct proportion to the number of news outlets reporting on the bank’s internal data. HSBC could not shut off The Guardian from publication, as it first attempted, because so many others would also be publishing at the same time. The upper hand seems to be with the successful collaboration that stretches around the world and does its reporting on a scale that finally matches the scale of the global enterprise.
  11. A joint deadline is sacrosanct
    One of the great strengths of any collaboration is joint publication on a single day or a single weekend. Dozens of news outlets in many countries begin telling a more or less single story; interviews and excerpts are then published and aired on hundreds of other news outlets over the following days. In a global collaboration, picking a single date and time is not an easy exercise. Every country seems to have a particular day of the week that is its preferred release time for big projects. Sunday is the big release day in the U.S., but Saturday is preferred in Germany; other nations want to avoid a weekend release altogether. The more concentrated an investigative project’s release is, the bigger the potential impact. With dozens of news organizations involved there are always some that won’t want to hold back from publishing their own work to wait for the group, and others that are just not ready to publish by the time the rest of the group is set to go. These matters take endless negotiation, hence the need for central decision making from an editorial core (see number one above).
  12. Credits and awards are shareable commodities
    Letting go and sharing credit do not come easily to most competitive news organizations or journalists. But these attributes are required in a big global collaboration. The work was been done by many so the credit needs to be shared. Just as each news organization does its own work, each is free to enter that work in its own country for as many awards and as much recognition by its peers as possible. And ICIJ members do win plenty of awards in their home countries. “I’m happy to tell you that I got the biggest journalism award in Finland,” wrote Minna Knus-Galan, for example, a Finnish TV reporter and ICIJ member. She won recognition last March for her ICIJ cross-border investigative work, especially the Luxembourg Leaks project. She said the award gave her a chance to speak about the overall success of ICIJ. One requirement for each member is to give credit to the editorial core, the place where they obtained their data and documents to begin with. This may seem simple, but sharing credit with dozens or even hundreds of others can be complicated in its details and execution.

Read the full paper.

Posted in Data Journalism, Journalism, Posts from feeds Tagged with: , , , , , , ,