Michael Braga had reached that point every reporter dreads: He was floundering, without a story idea, and was miserable as a result. It was early 2014, and he and Anthony Cormier, then investigations editor at the Sarasota Herald-Tribune, were coming off a yearlong series chronicling the collapse of dozens of banks in Florida. And the two-man investigative team knew it was important to get working on another big story—soon.

The Herald-Tribune has been committed to deep investigative work for years, winning a Pulitzer in 2011 for reporter Paige St. John’s examination of Florida’s troubled property insurance system. But like newspapers everywhere, by 2014, the Sarasota newsroom was down roughly 40 percent; daily circulation was, too. And neither Braga nor Cormier had the luxury to just sit around. Braga was responsible for writing a real estate blog that consumed two days of his week and was a prolific writer of briefs, penning hundreds of them over the course of the year. Cormier had other stories to edit. Together, while they toiled, they searched—and searched some more—for the next big story. “We didn’t know what to do,” Braga says. “We didn’t have another big idea.”

Four months slipped away, until they hit on something new. Years earlier, the paper had acquired a database of crimes committed throughout the state. “Every crime,” Braga says. “It’s a giant database. If we printed it all out, it would run for 38 miles.”

Braga and Cormier wanted to analyze the database for trends, believing it would be fertile ground for a new story. There was just one problem. “I couldn’t make heads or tails of it,” Cormier says. “It was just so massive, just terabytes of data. I’m a good data reporter. But I’m not a programmer, and we needed programming skills to unlock this data.”

Late one night that spring, Cormier sent a text to an unlikely person: Chris Davis, his competitor, deputy managing editor for investigations at the Tampa Bay Times, a paper 40 miles to the north, with three-and-a-half times the circulation and a nine-person investigative team, including reporters, editors, and researchers. Davis had once overseen investigations at the Herald-Tribune, before moving in 2011 to the Times, then called the St. Petersburg Times. Sarasota had acquired the massive crime data under his watch. And Braga and Cormier thought it only right to let him know they were planning to use it. But this wasn’t just a courtesy call. They wanted Davis’s help.

What followed was an 18-month collaboration between two would-be rivals. For months, the team chased a thesis that led nowhere, until Davis finally killed the approach. Still, the Tampa Bay editor refused to let the two papers walk away. There was a story in the data; Davis was sure of it.

Finally, they found it: Florida courts were sending thousands of patients every year into state-funded mental hospitals while lawmakers slashed their budgets by $100 million. The result, the two papers jointly declared, “mental patients are warehoused, cared for by startlingly few trained workers, and living in a violent environment that has led to the death and injury of patients and staff.”

The story “Insane. Invisible. In Danger.”—written by the Tampa Bay Times’s Leonora LaPeter Anton, along with and Braga—won this year’s Pulitzer Prize for Investigative Reporting. And what’s interesting about it isn’t just the collaboration (that’s happened before) or the Pulitzer (that’s happened with collaborations before, too), but the players themselves: two papers, in each other’s backyards, willing to get creative in the name of the story, and what that willingness says about the broader journalistic landscape. “Ten years ago, I don’t think these two papers would have done this,” says Cormier, who had moved over to the Times by the time the story was published. “We would have been competing to do it on our own. But the state of journalism forced us to pool our resources to pull it off.”

In order to do big stories in difficult financial times, news outfits—especially smaller organizations—are increasingly going against old instincts: cooperating instead of competing and conceding turf instead of fighting for it, while at the same time doubling down on the old bet that readers, even today’s distracted, mobile readers, want in-depth investigative reporting. Watchdog stories like “Insane. Invisible. In Danger.” remain one of the only ways citizens can hold government—big and small—in check, make elected leaders and corporations accountable for their decisions, and stave off abuses of power.

To do big stories in difficult financial times, news outfits—especially smaller organizations—are increasingly  cooperating instead of competing and conceding turf instead of fighting for it

“Investigative stories are not celebrity news,” says Mark Katches, editor of The Oregonian and vice president of content for the Oregonian Media Group, who’s led investigative teams that have won two Pulitzers and been finalists five other times. “They’re not crazy trending stories. They’re not viral, typically. You’ll see outliers. But they are not going to be the massive traffic drivers. History shows that.”

And yet, Katches and others believe that, in order to survive, traditional newspapers, nonprofit news organizations, and for-profit websites have to focus on a metric not tracked by comScore: the impact, beyond the numbers of how a story resonates in a community, or forces leaders to act, or, at times, maybe even changes the world. “That’s the number I need,” says Mitch Pugh, executive editor of The Post and Courier in Charleston, South Carolina. “It’s great to have page views and unique views and everything. But how do I measure the wider impact of the journalism?”

It wasn’t too long ago that it was common for small papers like the Sarasota Herald-Tribune and Tampa Bay Times to write stories like “Insane. Invisible. In Danger.”—and to win Pulitzer Prizes for them. Between 1917, when the awards began, and 1999, small papers routinely took home the award for public service. (The award for investigative reporting wasn’t created until 1984.) But staff cutbacks have made it harder for smaller outfits to dedicate resources to big investigations. And this void has been especially felt in communities far away from the last big American newsrooms, where the national papers rarely visit and the local paper doesn’t carry as much weight as it once did. “These mid-sized to large metro papers were beacons in their community,” says Brendan McCarthy, a former reporter for The Times-Picayune in New Orleans and a 2009 Pulitzer finalist for local reporting. “And that’s changed a lot.”

One thing that hasn’t changed is the speech top editors everywhere give reporters when they’re hired: They are going to be committed to watchdog journalism; they are going to ask tough questions and do big stories. But it’s a speech that cynical reporters, who have attended five-too-many send-offs for laid-off colleagues, take less seriously today, in a time of shrinking ad revenue, circulation, and staff.

Between 2005 and 2015, American newsrooms have bled staff at alarming, unprecedented rates. According to the American Society of News Editors’ Newsroom Employment Census, the total work force fell by nearly 40 percent in the last decade—from 54,100 to 32,900—leaving a lot of empty cubicles. During the same time, newsrooms, working with smaller staffs, are doing more, producing not only print editions, but digital stories, videos, blogs, chats, and podcasts. That means in-depth, long-form stories are “harder to manage,” according to Susan Ellerbach, executive editor of the Tulsa World: “If you have four to five people on larger projects, and you’ve taken them out of the daily run, that cuts back on your staff trying to fill the newspaper and updating the website to keep it fresh.”

The sweeping changes have affected the way newspapers cover just about everything, but perhaps most troubling is the way they cover government. A Pew Research Center study found that the number of full-time newspaper reporters assigned to U.S. statehouses declined by 35 percent between 2003 and 2014. The vast majority of papers and local television news stations don’t have a single statehouse beat writer these days, either full time or part time. The same cutbacks have gutted local beat reporting. And that worries Amy Pyle, managing editor at the Center for Investigative Reporting, the California-based nonprofit news organization. She wonders what stories news agencies are losing—or, indeed, missing altogether—in this historic downsizing of old-fashioned beat reporting. Without these cutbacks, would reporters have uncovered police abuses in Chicago earlier? Or the tainted water in Flint, Mich.? And if so, what’s the solution? “I don’t know,” Pyle says. “I think we’re trying to be part of it. I think all of us, collectively, are trying to be part of it.”

At South Carolina’s Post and Courier, Pugh has tried to strike a balance. During moments of breaking news, he shares a simple directive with the newsroom: “Two sentences!” He wants two sentences up on The Post and Courier site; the bigger story can come later. But at the same time, the paper has remained committed to its watchdog team—a three-person outfit when Pugh arrived.

In early 2014, as Braga and Cormier embarked on their new investigation in Florida, Glenn Hill, watchdog and public service editor at The Post and Courier, was doing the same thing up in Charleston. A few months earlier, the paper had received a press release from the Violence Policy Center, a Washington, D.C.-based advocacy group, declaring that South Carolina ranked first in the nation for men killing women, the tenth time in the last decade that the state had made the top 10. “We started talking about that we were No. 1 that year, the third time we were No. 1,” Hill recalls. “And I think we were getting ready to do our annual story. Probably 15-20 inches. Run on the front page. Call the usual suspects. And we started having a deeper conversation: Why do we keep ending up on this list?”

Hill phoned in to the conference call the Violence Policy Center had set up to discuss the numbers in South Carolina. “I was the only one who ended up calling in,” he says. No other reporters from other state papers were on the line. And at the state capitol, there was no outrage, not even staged anger for the press. “We’re No. 1 in the nation for women killed by men—and we’re there for years and years and years—and there doesn’t seem to be any great effort to do anything about it,” says Hill. “Meanwhile, women were dying by the dozen, one every 12 days. And the more we looked into it, we found the state did almost nothing to track these deaths.”

The Post and Courier set out to track them instead, with a small but focused team. It was Hill, and his two reporters, plus a third assigned by Pugh, with high-level editing and database training provided by the Center for Investigative Reporting (CIR). A CIR boot camp helped Charleston staffers better understand how to find the documents—and numbers—they needed to tell this story. But given their small staff and the big ambitions Hill and his team had for the story, they organized themselves differently than in the past.

From the start, they developed a story memo, clearly stating the idea, and held meetings attended not only by the reporters working on the story, but by the top editors, the interactive folks who might shoot video, and even the page designers. “Project reporters, traditionally, work in a separate office and don’t say a lot about what they’re working on, and sort of spring it on everyone at the last minute,” Hill says. “And what we’re trying to do now is get beyond that somewhat.”

In recent years, nonprofits like KyCIR have popped up across the country, filling in for flagging newspapers and freeing up reporters from financial constraints

What they learned, Hill says, from their seven-part series, “Till Death Do Us Part,” was that it’s OK to share plans and even early drafts. “It’s getting comfortable with sharing earlier,” according to Hill. And not just with people in-house. The Post and Courier team assigned to the story—small and nimble, by necessity—also got outside feedback from CIR. And together, it worked. The Post and Courier series not only won the Pulitzer for Public Service, it forced state lawmakers to pass sweeping domestic violence reforms, including gun bans for repeat abusers, in less than a year.

Three years ago, former Times-Picayune reporter Brendan McCarthy left New Orleans to help launch the Kentucky Center for Investigative Reporting (KyCIR), a nonprofit news venture backed by Louisville Public Media, which operates three public radio stations, including WFPL, Louisville’s NPR affiliate. McCarthy, who started his career in newspapers, saw the effort as a way to do work he felt was important and not happening as often as it once did: monitoring power and holding decision-makers accountable. In recent years, nonprofits like KyCIR have popped up across the country, filling in for flagging newspapers and freeing up reporters from financial constraints. But the ventures often come with their own set of challenges—finding donors, instead of advertisers, and starting over, with almost nothing.

“Literally, we showed up in an empty room,” McCarthy recalls, “a small, windowless, white-walled, sanitized room.” The good news was, they didn’t plan to spend much time there. McCarthy envisioned them traveling the Commonwealth and meeting with newspaper editors to introduce themselves and the meaningful investigative work they hoped to do—and give to the papers for free. But the road trip was much shorter than he imagined: “Early on, a lot of people looked at us skeptically. Who are these folks? There were dozens of examples where no one called me back.”

They’re calling now. In the three years since its launch, KyCIR’s journalism has run in newspapers—and aired on radio stations—across the state, collecting national Edward R. Murrow and Investigative Reporters and Editors awards. Its audience doubled to nearly 3.6 million impressions in 2015. “Our commodity is great investigative journalism that we hope makes an impact,” McCarthy says.

Great investigative journalism is a product other nontraditional investigative centers are increasingly making, too. Some, like the New England Center for Investigative Reporting, formed in 2009, are nonprofits. At least one, The Frontier, in Tulsa, is selling subscriptions: $30 a month for full access to all its local investigative content—content that The Frontier’s founder, Robert E. Lorton III, says he doesn’t get from his local traditional sources anymore. “What we’re seeing is shorter stories, fewer stories, and not going after watchdog stories,” says Lorton, former publisher of the Tulsa World. “And ultimately, that’s the mission of any newspaper: keeping people honest, asking tough questions, along with trying to cover all the surface stuff.”

The venture isn’t yet profitable; the website has about 500 subscribers. “But we’re doing a lot of good for this community,” says Ziva Branstetter, a former investigative reporter for the World who now leads Lorton’s team. And there’s evidence—nationally, anyway—that alternative news outfits, committed to deep investigative work, aren’t so alternative anymore. In recent years, ProPublica, the Center for Public Integrity, and InsideClimate News have each won at least one Pulitzer for their work. ProPublica was a winner again this year, sharing the prize with another nonprofit: The Marshall Project, a watchdog unit dedicated to covering the American criminal justice system.

Bill Keller, former executive editor of The New York Times, helped launch the effort in late 2014, and one of his first hires was Ken Armstrong, a longtime Seattle Times investigative reporter and winner of the 2012 Pulitzer Prize for Investigative Reporting. On Armstrong’s list of story ideas at the time was something Keller began to describe as Armstrong’s “white whale”: He wanted to tell the story of 2008 rape victim in Washington state, coerced by police into saying she fabricated the tale, and then charged with a gross misdemeanor for that apparent lie, while the rapist went free, assaulting others, until police arrested him, in Colorado, in 2011.

It was a story that required both hard-to-get documents and almost impossible interviews—with police, the victim, and the rapist, too. Armstrong estimates he must have had more than two dozen conversations with the victim’s attorney before she agreed to talk. But by the summer of 2015, he had managed to interview the woman—Marie, they called her—and other key characters in Washington. The story was coming together. And that’s when Keller got an e-mail from Joe Sexton, a ProPublica editor. The subject line mentioned a “delicate” question. “Bill,” the short note began, “I think we may be working on the same story.”

One of Sexton’s reporters, T. Christian Miller, had been investigating police failures in sexual assault cases. A former prosecutor told him about this shocking arrest in Colorado. Miller had interviewed the Colorado detective who had made the arrest and worked his way to Marie’s attorney in Washington. And that’s when he made the discovery: The Marshall Project was investigating the story, too. “You’re just sort of instinctively competitive in a situation like that,” Keller says. “And the first thing I thought was, Are we far enough along that we can rush this thing into print, so that we’re not behind ProPublica?”

He forwarded Sexton’s e-mail to Armstrong with a brief note of his own: “Oh, shit.”

Armstrong didn’t want to rush the story, but he didn’t want to get beat, either. It was Miller who suggested they join forces. And once he did, Armstrong says, “it didn’t take long for me to see the enormous benefits of working on this together.”

The result was a 12,000-word story, published last December, constructed with public records and interviews, and written in a polished narrative, showing just how—and why—the system had failed. On The Marshall Project site, the story generated about a half million visits. On the ProPublica site, it was even larger: 800,000. And, according to the metrics, readers lingered. They actually read, spending an average of 15 minutes on the page, a lifetime in the digital era. The Pulitzer judges read, too, awarding it the prize last spring for explanatory reporting.

A new study suggests Armstrong and Miller’s rape story may not be an outlier. After analyzing more than 400,000 stories at 55 publications, Tom Rosenstiel, executive director of the American Press Institute, concluded that stories averaging 1,200 words drove 23 percent more engagement, defined, generally speaking, as page views, time spent, and sharing. These “long-form” stories lifted page views (up 11 percent), sharing (up 45 percent), and reading time (up 36 percent). And so-called “major enterprise” stories did even better, generating 83 percent more page views and double the sharing activity.

Still, Keller concedes, there are challenges ahead. The Marshall Project, despite its recent Pulitzer, still has just a fraction of the budget Keller once enjoyed at The New York Times: $4.6 million, compared to about $200 million. And while reporters like Armstrong are generating great stories, the work isn’t reaching most pockets of America, especially the smaller markets. “In order to survive we have to convince the people who support us that we are making a difference,” Keller says. “One measure—and by no means the most reliable measure, but it’s something that funders look at—is how many people read the story and might have been moved by it?”

Keller says it’s “an imperfect metric,” but one that motivates nonprofits to partner with, say, The Washington Post, rather than a small daily, with limited reach. “It’s a real problem,” he says. “For us to justify investing a lot of time in collaboration with a local new organization, we have to be able to use that locality as microcosm for something larger that’s happening in the country. You can’t always do that.”

And sometimes, small organizations choose not to work with The Marshall Project. Faced with the offer of a well-reported story (copyedited, legally vetted and ready to run, for free) they sometimes pass—ironically, for the same reasons they are no longer able to do those stories themselves. “Offering a 2,500-word piece to a small daily newspaper, that’s asking them to swallow a lot,” Keller says. “And, of course, to be responsible, they have to do their own edit, and convince themselves that the story is the way it should be, and do a copyedit. And they’re just stretched too thin.”

George Stanley, editor of the Milwaukee Journal Sentinel, knows the feeling. As recently as the late 1990s, the paper had a news staff hovering around 300. Today, it’s more like 120.

But one thing at the Journal Sentinel hasn’t changed: The paper remains committed to watchdog journalism. Its investigative team has 11.5 staffers, more than it had a decade ago and an impressive ratio, given what’s happened at the paper overall. Stanley says he has another two reporters, unofficially on the team, working primarily on major investigative and explanatory projects. And the paper has created a culture where everyone can do big work, if they stumble upon it. All they need to do, Stanley says, is let an editor know that they’ve got something interesting, something worth a look. “I don’t care if it’s a photographer, or a graphic artist, or a reporter, we’re never going to take that story away from you,” Stanley says. “We’ll build the infrastructure around you. And that encourages everybody to look for big stories.” Exactly the kind of stories the Pulitzers were established to honor.