How to create a culture and structure for innovation
It’s an old line, but David Skok said it bears repeating:
Culture eats strategy for breakfast”
Skok is digital advisor to the editor-in-chief of the Boston Globe, and a former Nieman Fellow. He spent his year at Harvard studying and collaborating with the creator of disruptive innovation theory, Clayton Christensen. They co-authored an article about disruption in the news industry.
Culture is a focal point of any discussion about innovation. In Christensen’s disruptive innovation theory, a culture manifests itself as RPP: resources, processes and priorities. “A culture is made up of all of those things, ” Skok said. “You first have to have the priority set from on high.” He continued:
You have to align the resources consistent with those priorities, and establish the processes, and that’s all very abstract, but the basic idea is: somebody gets to work every morning in the newsroom. They turn on their computer. What are they doing when they turn on the computer? What is going on in their head, what is their objective? And if you actually can’t answer that question in the way that you want to answer it, then your culture is messed up.
The danger, according to Skok, is that leaders simply tell people that the goal is to be innovative, without changing the “structure or the tasks that people do to allow that to happen.”
“Then it’s just shouting empty words,” he said.
“Leaders cannot simply mandate a new culture,” wrote Brown and Groves in their paper. “Organizations must develop new routines that fit in the context of the existing culture and nudge members toward a culture that embraces innovation.”
Research into organizational change has identified an organization’s existing culture as one of the key barriers to making real change, according to their paper.
“Culture is the ultimate source of most of the defensive mechanisms that block organizational change and prevent learning from occurring,” they wrote.
Traditional organizations have been doing things a certain way for a long time, and the way people work and interact are a reflection of this. The authors of The New York Times innovation report — which the Nieman Journalism Lab called “an astonishing look inside the cultural change still needed in the shift to digital” — highlighted the fact that the front page of the paper is still a dominant focus for the organization:
Our internal fixation on it can be unhealthy, disproportionate and ultimately counterproductive. Just think about how many points in our day are still oriented around A1 — from the 10 a.m. meeting to the summaries that reporters file in the early afternoon to the editing time that goes into those summaries to the moment the verdict is rendered at 4:30. In Washington, there’s even an email that goes out to the entire bureau alerting everyone which six stories made it. That doesn’t sound to me like a newsroom that’s thinking enough about the web.
Those elements of process all reinforce a print-first culture. Shifting away from that requires re-examining all aspects of the organization. Groves, a professor at Drury University who co-authored the “Lean Newsroom” paper with Brown, said that culture is “the core of the organization itself; it’s the foundation of how the organization functions and accomplishes its mission.”
He cautioned newsroom leaders against thinking they can simply create a new culture.
“Culture is a shared phenomenon, and although leaders play a key role in shaping it, they rarely create it completely by themselves, especially in a well-established organization whose underlying assumptions are formed through years of tradition and success,” he said.
The starting point, according to Groves, is to have a clear understanding of your existing culture, something he says leaders often lack.
“I bet nine times out of 10 they can’t tell you [what their culture is], or they will say it’s a culture that won’t change,” he said. “That is a cop out.”
The process of defining your existing culture involves things such as understanding the work and actions that are celebrated and rewarded, and which things you do not do well, according to Groves.
“People don’t like to go through that self-reflective process,” he said. “The other issue you run into is that somebody will say, ‘Oh look at what FiveThirtyEight is doing, why can’t we do that?’ Well, you’re not FiveThirtyEight. That’s not going to be your answer. What you have to do, the hard thing about this, is that you have to find your answer. And your answer [comes from] understanding who you are and what you’re capable of doing.”
Their paper advocates that leaders think carefully about the message their actions send, and work to reinforce aspects of the culture they want.
“Leaders seeking to evolve their culture toward greater iteration and learning need to examine carefully what is rewarded and punished, and recognize the critical role they play not only in what they say but in every action they take, no matter how subtle,” they wrote. “Staffers are quick to notice not only the more obvious internal awards or raises but also how attention and simple praise are distributed.”
One key aspect of a culture that nurtures innovation is collaboration: across departments and teams, and up and down the organization. It was mentioned by everyone interviewed for this study.
Vox’s Brundrett said a culture of collaboration flows from mutual trust.
“The big lesson for me is it’s really just a big issue around creating mutual trust and respect between teams,” he said. “If you’re working together it’s really about understanding where they are coming from and their background and where the hard work is, and what’s valuable that they bring to it. And when you do that, instead of getting the editorial team telling the product team what to build, what you have is people making things together, having ownership together and trusting each other. And much better things come out of that.”
Coupled with that sense of collaboration, and mutual trust, must be a commitment to create a culture where people are given ownership and the ability to act.
“I’m a strong believer that performance management fundamentally is about giving people ownership over their own product and what it is that they’re trying to do,” Skok said. “The more you can own that product or feel that you have ownership over your ability to influence it, the more empowered you are to go forward and do things that may be innovative.”
The result, he said, is that innovation as a topic never comes up. It just happens.
“We don’t want to just innovate by creating new products, we want to innovate by creating the culture that will allow for that innovation space to happen,” he said. “So it’s kind of like we’re putting together the pieces of a Rubik’s Cube so that they fit, and then once they fit, then we hope it’s … part of our processes every day, that we’re always innovating without even having to say we are.”
We don’t want to just innovate by creating new products, we want to innovate by creating the culture that will allow for that innovation space to happen.”
Innovate with organizational structure
Vox’s Brundrett confessed that he is currently obsessed with org charts. But his view of this traditional organizational artifact is very different: he sees it as a design document.
Vox is growing rapidly in headcount and in the verticals and offerings it has for readers and advertisers. Amidst this growth, Brundrett and other managers are focused on designing an organizational structure that continues to foster innovation, rather than one that exists to enable administrative efficiency.
“So much of what we spend time thinking about and tweaking and working against is the inertia of [what happens] as you get larger,” he said. “You end up trying to organize yourself in ways that are easier to manage and administer, but that aren’t necessarily the most productive for the way that you create things, and the way you get people to collaborate.”
He said it’s positive that many news organizations are now putting developers in the newsroom. But this initiative can go awry when those developers don’t have access to technology resources that enable them to actually build and deploy things, according to Brundrett. The structure isn’t in place.
“Because they are not in the technology part of the organization. they are stymied by, and have to work around, not being able to plug into the tech resources and platform,” he said. “We’ve been trying this experiment where we have developers and designers that are on the product team and are embedded with the newsroom, and they have full access to the platform. They are sitting every day in meetings, pitching stories and building things and taking bylines, but the idea is that they are plugged into the infrastructure and there is not a kind of an ‘us versus them’ piece.”
Brundrett said one sign of whether groups are integrated is the language used to refer to different parts of the organization. When people talk about the “editorial side” or the “product side,” it’s a giveaway that they are not one unit.
“That vocabulary is super telling because it’s like, ‘What do you mean by side? That side of the room? Or that side of the [field]?'” he said. “The key is to try to get creative and to try some experiments and innovate around some organizational pieces and see what pressure that puts around the culture and communication.”
One sign of whether groups are integrated is the language used to refer to different parts of the organization. When people talk about the ‘editorial side’ or the ‘product side,’ it’s a giveaway that they are not one unit.”
Innovation flows from a willingness to experiment and find the right structure and culture that result in new things being created, according to Brundrett.
“That’s much more innovative to me than just building the most innovative technology, because the first one leads to second one,” he said. “You look at any of these organizations that are doing a good job with innovating in design or media and it usually is the case that their organization and culture is set up in a way that is not the norm.”
Medium, the publishing platform co-founded by Twitter’s Evan Williams, adheres to a unique organizational approach called Holocracy. It’s often referred to as management without managers. (The online retailer Zappos.com has also implemented Holocracy.)
Holocracy.org, the website of the organization that administers the Holocracy trademark and training offerings, describes Holocracy this way:
Unlike conventional top-down or progressive bottom-up approaches, it integrates the benefits of both without relying on parental heroic leaders. Everyone becomes a leader of their roles and a follower of others’, processing tensions with real authority and real responsibility, through dynamic governance and transparent operations.
Kate Lee, a senior editor at Medium, said “Holocracy is very much a culture of autonomy with accountability.”
“It’s not a consensus-based culture — it is biased toward action but also being accountable for what you do, and understanding the results of whatever you do,” she said.
Lee said Williams, Medium’s CEO, implemented Holocracy because it empowers individuals and teams inside the organizations.
“It’s really about distributing authority to different people who then have the autonomy to lead and to do their jobs,” she said. “There’s no question that Ev is our co-founder and CEO, but part of the reason we have this corporate structure is he doesn’t feel like he has to be the heroic CEO who can solve everyone’s problems. We all feel empowered to solve our own problems.”
Evaluate and reengineer processes
Two years ago, Chisenhall saw she needed to change the way her team at the Wichita Eagle worked. Otherwise, they would never begin to adapt to the digital future. In the hope of sparking change and innovation, she tore up the old model for working hours and jobs.
The old workflow saw the vast majority of content delivered late in the afternoon. That was the print mentality. Instead, reporters now work to have content published starting early in the morning, and throughout the day. The organization implemented a new schedule for copy editors to ensure they were staggered during day, rather than having them all on duty in the afternoon and evening.
“We examined our process and how we were doing things and challenged our thinking,” she said. “It involved everything from our processes to the way you think to the vocabulary you use.”
Tomasik underwent the same process at his Florida news organization: they changed the work schedules and meeting times, and pushed people to change the work they did, when they did it, and how they did it.
An example of the need for traditional media organizations to restructure their processes and workflows was captured in The New York Times innovation report:
Stories are typically filed late in the day. Our mobile apps are organized by print sections. Desks meticulously lay out their sections but spend little time thinking about social strategies. Traditional reporting skills are the top priority in hiring and promotion. The habits and traditions built over a century and a half of putting out the paper are a powerful, conservative force as we transition to digital — none more so than the gravitational pull of Page One. Some of our traditional competitors have aggressively reorganized around a digital-first rather than a print-first schedule. The health and profitability of our print paper means we don’t yet need to follow them down this path.
It’s essential to create new processes for meetings and information-sharing as a way to spark collaboration and reinforce new roles and practices, according to nearly everyone interviewed for this study.
At TIME, one new process is a bi-weekly meeting that brings together all key stakeholders. It is aligned with their implementation of the Agile development methodology, which sees the product team engage in code sprints to launch and iterate new and existing products. (A code sprint is a set time during which the development team works to evolve a product based on specific requirements.)
“That’s where we have tech, product, editorial, all stakeholders represented, talking through the various products and what’s coming in the pipeline,” said Callie Schweitzer, the editorial director of audience strategy for TIME and Time Inc.
Breaking down silos and creating processes that bring people together is a key to fostering innovation, according to Schweitzer. (This aspect is examined in more detail in the next section.)
“I think that in the past, a lot of organizations have functioned where one department would be working on one thing and another department would be working on something else in parallel and we just don’t even let that happen here,” Schweitzer said. “We now have every key stakeholder at the table talking about things and figuring out what a project will entail and what it will mean for every side of the business.”
Tom Meagher is another person thinking about redesigning newsroom workflow and processes. Meagher was the head of the interactive team for Digital First Media’s now-defunct Thunderdome newsroom, which was a centralized desk serving newspapers all over the United States.
At the SRCCON conference, he led a session entitled “Before we can fix the news, we need to redesign the newsroom.” The goal was to “talk about how we can redesign the newsroom itself to upend our clunky workflows pegged to Sunday’s newspaper instead of today’s news online.”
In his recap of the session, Meagher outlined the reality of the siloed newsroom experience that still keeps developers and product people separate from editorial and IT:
Until this spring, I worked for a major U.S. newspaper chain that operated 75 papers in 17 states. The company had zero developers in its newsrooms working on editorial projects. At my paper before that, we had zero developers in our newsroom, and no one in the company overtly doing news development. The status quo in both of those companies was that my team had to wage a guerrilla war on the IT departments and the product and CMS divisions to do the kind of journalism we know is important and to get our jobs done.
Participants in his session came up with new structures for news organizations, and imagined the related workflow. One model was for a single-subject news website. The below image shows the staffing structure of the site. Rather than present the team in a typical org chart, they use concentric circles to show that each group is connected. For example, reporters and editors are in the same rung as developers and photographers because they will collaborate on projects:
Here’s how they envisioned the collaborative workflow for this organization (which was called “The Score”):
At daily meetings, all department heads and project managers would pitch stories and coordinate staffing. For each project, ad-hoc teams with specialists in each medium would be spun up, and they would co-locate for the duration of the project. They’d hope this could make journalists more tech savvy while forcing coders to be more editorially minded. For the developers, the bar would be set very high for documentation and reusability. The idea is for all of the parts of the newsroom to function as an orchestra, making beautiful music together with the project manager as the conductor (thus “The Score”).
The Score was imagined as a newsroom where projects start and are then continued or curtailed based on set goals. Some projects would conceivably fulfill their original vision, while others might evolve in different ways, or need to be shut down due to a lack of traction. Then those involved would apply their lessons to the next project.
This aligns with Brown’s and Groves’s finding that organizations “must … engage in a continuous process of problem solving, in which failure is recognized as valuable information.” Even if a project doesn’t meet its goals, it can still provides valuable information and lessons to the organization.
The message, again, is that iteration and constant refinement are not just for product development — they must become ingrained in the organization’s overall culture and processes.
“Importantly, iterative processes aren’t just relevant to creating new products like an app or vertical,” write Brown and Groves in their paper. “These processes can be used to allocate resources more efficiently in newsrooms with smaller-than-desired staff sizes on stories and projects.”
Create multidisciplinary teams that collaborate
When Quartz, the fast-growing global business website from Atlantic Media, announced that it had created a chart building tool to enable reporters to quickly generate visually appealing charts, some of the writers and editors at its sister publications were a bit jealous.
Kimberly Lau of The Atlantic Digital said people asked why they didn’t also have an easy chart builder, and if one could be built.
Lau said a frustrated product manager on her team told her, “we built a chart tool a year and half ago and nobody used it.”
Why was the Quartz tool so popular with its writers, and the one built for The Atlantic gathering dust?
Lau said collaboration was the key ingredient in Quartz’s success. The Quartz tool was built as a result of a discussion between its editorial and product teams. In fact, the tool itself was built by Quartz journalist David Yanofsky, who describes himself as “a professional journalist, trained designer, [and] self-taught programmer…”
Yanofsky saw the need for the tool on several levels, including the site’s visual identity.
“We had ugly charts on a beautiful site,” he said.
So he built a tool to address that need. His colleagues at Atlantic Media, however, didn’t build their tool the same way.
“Our product team thought it was a great idea but didn’t have the editorial advocate … that was willing to be a part of that conversation,” Lau said. To her, the Quartz Chart Builder is a true example of innovation.
“It’s not just about [something] new, because new is definitely exciting and nice — but if it’s not adopted, it doesn’t really matter,” she said. “What we’re looking for is adopted innovation.”
Chart Builder also succeeded, Lau believes, because the collaborative process that included the right mix of people. The developer understood and empathized with the needs of his target user. Indeed, he was the target user.
If it’s not adopted, it doesn’t really matter. What we’re looking for is adopted innovation.”
The same principle of interdisciplinary teams applies to products built for wider audiences.
In 2007, Bill Adair was the Washington bureau chief for the St. Petersburg Times (now the Tampa Bay Times). After covering the White House in 2003 and 2004, he felt guilty about the lack of fact checking being done about politicians’ claims.
“I covered the Bush campaign and felt like after that campaign I had not done any fact checking, and that I should have,” he said.
Adair told Times Editor Neil Brown he wanted to launch a fact-checking website in time for the 2008 campaign. Brown agreed to let him look into the idea. After he created a prototype, Brown signed off on building the fact-checking site. Adair then pulled together a team of people with diverse backgrounds to build it and write the first fact-checks.
“How can you do something like this at a legacy news organization? I think you have to create small teams and sort of break away from the mothership,” he said. “And that’s what we did.”
Adair was joined by Angie Holan, who at the time was working in the newspaper’s library.
“She was a librarian with great research skills, but she was also a very talented writer,” Adair said. “She had a lot of digital skills and so she was ideal as we evolved because she could do a little bit of everything.”
Holan is today the editor of PolitiFact.
Matt Waite, another early team member, was a special projects reporter who had recently transitioned into more of a technical role. He suggested that PolitiFact apply the emerging concept of structured journalism, which Adrian Holovaty had blogged about in 2006. That’s why today’s PolitiFact enables you to see all of the Truth-O-Meter ratings for a given politician. It was built from day one to be a database of fact checks, rather than just a collection of articles.
Adair said this interdisciplinary team was key to the creation and ultimate success of PolitiFact, which won a Pulitzer Prize in 2009.
How can you do something like this at a legacy news organization? I think you have to create small teams and sort of break away from the mothership.”
The next element of team design most-cited by experts relates to their size. Over and over again, people interviewed for this study came back to the same rule: the team should be able to share a meal with nothing more than two pizzas.
The two-pizza rule is attributed to Amazon.com founder Jeff Bezos. Bezos “wanted a decentralized, even disorganized company where independent ideas would prevail over groupthink,” according to The Wall Street Journal.
The result was a lot of smaller task forces with five to seven people, who tackled a challenge together.
At Russ Media, they create small teams from different disciplines, and they are rigorous about giving teams autonomy.
“One of the most basic principles is that we don’t try to grow our teams too much,” Riedmann said. “We tend to organize everything in very small teams and a very federal structure, and they are asked to work together, but they can make their decisions in a very decentralized way.”
At Vox Media, teams are formed and disbanded to meet specific projects that align with priorities. Brundrett’s product team has just over 60 people, but they will join up with others from editorial, sales and other departments to get the right mix for a given project.
At the time we spoke, Brundrett’s product group had 10 active initiatives running with various sizes of teams. These initiatives included the ongoing evolution of the recently-launched Vox.com site, and the creation of a custom ad platform.
“We organize into small, cross-functional multidisciplinary teams that are organized around products and users, not around our own organizational structure,” he said. “They are organized often times around initiatives and they scale up and down depending on how ambitious the thing is we’re taking up.”
This approach is the opposite of how media organizations were structured in the past.
“Our policies are children of the industrial age, not the digital age,” Eric Newton wrote in “Searchlights and Sunglasses.” “They often block innovation and the creation of new journalism.”
The old model created silos. There was, as Brundrett said, the “editorial side” and “the business side.”
“As a result of that you had an organization that has metastasized all these different siloed pieces of the organization,” said Skok of the Boston Globe. “And in many ways, you know, they all had competing priorities … it generally had to be up to the publisher or some senior leader to get the head of engineering, and the head of editorial, the editor, in a room and say, ‘Well, this is what we’re doing.’ ”
Today, organizations that are effecting change and pursuing innovation are smashing silos and creating smaller, more nimble teams. These teams are able to pursue initiatives without encountering constant barriers.
Schweitzer’s role at TIME straddles the newsroom, consumer marketing, product and other departments. Increasingly, she said, the entire TIME organization is working as interdisciplinary units.
“On my team, we work and sit in editorial, but we work with IT, with consumer marketing, audience development, with business development — I mean everyone you can think of,” she said.
Today, organizations that are effecting change and pursuing innovation are smashing silos and creating smaller, more nimble teams. These teams are able to pursue initiatives without encountering constant barriers.”
Teams must also be built in such a way that no single person becomes a bottleneck, or the sole entity able to push things forward, according to Amy Webb.
She told a cautionary story about a whiteboard she saw at ABC News years ago.
“ABC had the most immense whiteboard I’d eve seen in my life,” she said. “That whiteboard was completely full from ceiling to the floor with regular handwriting- sized letters. It was full. There were a thousand ideas on that whiteboard.”
The problem was the ideas never went anywhere, and it was because of the way they had structured their product team at the time. (Webb said her company is not currently advising ABC, and therefore can’t speak to its current process.)
“There had been a product manager who was responsible for collecting and prioritizing those ideas, however that person quit,” she said. “The new hire didn’t have that same organizational skillset. So the foundation fell apart.”
Another approach to teams is to create dedicated innovation units or labs.
E.W. Scripps has a small group of digital leaders that were brought in from outside of the news organization. They are called the Digital Solutions Group, and are charged with “aggressively developing new business lines in this increasingly digital time.” They work closely with the existing leadership team and units within the company
So far, this group has helped launch products including a paid weather app that draws on Scripps’ local expertise, and a portal for youth sports aimed at mothers, according to a report by NetNewsCheck.
However, experts and newsroom leaders contacted for this study cautioned against creating an innovation silo, or giving one person the responsibility for innovation. They emphasized the need to create a culture and structure that make innovation part of the daily operations of the organization. It’s everyone’s job.
“This is not about creating a title or an innovation team, because all that is doing is throwing existing staff and existing mindsets at tough challenges that need to be solved,” Webb said. “The solution to that problem is modeling the entire process in a very different way that is sometimes antithetical to how news is created. All the places that hired chief innovation officers or directors have created an additional layer of problems because what they have done is create yet another silo.”
Skok said an organization needs to ask itself why it’s considering a lab or dedicated innovation team.
“I would posit that if you are creating a lab, you’re probably doing it in the hopes that it will inform the rest of your business,” he said. “My suspicion is, if you’re creating that lab to be able to then inform and change the culture of the existing operation, it will be swallowed up by the existing operation every time.”
Another concern, he said, is a lab ends up being a sandbox where people try things that never really become integrated into the entire organization.
“Now, if you are looking to create a space where you can actually create a new business with its own priorities, processes and resources, that could then in a way disrupt the existing business, then you’re on to something,” he said, identifying the approach taken by Russ Media.
The Boston Globe in fact had a dedicated innovation lab. But the team was disbanded in 2014.
“One of the most difficult elements of innovation is making the hard decisions on good ideas that no longer make sense,” Skok said. “The Lab was a terrific concept but it was no longer part of our core operations, and we felt that we needed to make that our singular focus.”
Instead, the Globe decided to focus on establishing an organization-wide approach to innovation, according to Skok.