“For the last year, I’ve spent every working day trying to figure out where a high school kid was for an hour after school one day in 1999.”

On October 3, 2014, the first episode of a new podcast dropped. Serial quickly generated a tsunami of interest in true crime, and in podcasts as a way to reach previously untapped audiences. Before the show’s launch, Serial executive producer Julie Snyder told Nieman Lab its audience would “certainly be much smaller” than that of This American Life. It turns out that as of early 2024, Serial’s podcasts had been downloaded more than 743 million times.

The interest in true crime dates back hundreds of years, and crime has been a fixture of news coverage for nearly that long. Still, “the popularity of this podcast, I was unprepared for,” Serial host and executive producer Sarah Koenig told Nieman Storyboard in 2014. “I think a lot of that is the fact that it’s a crime. It’s a murder case.”

True crime is “a genre that encompasses as much as you could possibly think of,” said Sarah Weinman, the author of Scoundrel and The Real Lolita, editor of several crime anthologies, and writer of The New York Times Book Review’s crime and mystery column. She pointed to the most recent season of In the Dark, which focuses on the killing of 24 unarmed Iraqi civilians by U.S. marines in 2005. “Ten years ago, you would not have said that a season about the Haditha massacre counted as true crime,” she said.¹ Today, though, war crime is part of an ecosystem that stretches from Dateline and evening news segments to TikTok reels and NextDoor posts to local news startups and criminal justice reporting initiatives.

Ten years after the release of the first season of Serial, Nieman Lab set out to get some sense of what the crime reporting ecosystem looks like now, delving into how crime and journalism have intertwined over the past decade and how the reporting is changing.

“Let’s be honest: Crime coverage is terrible.” That’s what Tauhid Chappell and Mike Rispoli wrote in their Nieman Lab prediction in December 2020, six months after George Floyd was killed by police. “It’s racist, classist, fear-based clickbait masking as journalism. It creates lasting harm for the communities that newsrooms are supposed to serve. And because it so rarely meets the public’s needs, it’s almost never newsworthy.”

What’s changed since their prediction, titled “Defund the crime beat”? Even before 2020, some news outlets had already stopped running most mugshots and begun coordinated efforts to review people’s requests to have their names removed from old stories. After Floyd’s murder, some of those efforts accelerated. The Philadelphia Inquirer stopped automatically running mugshots in 2020. The Boston Globe, through its “Fresh Start” initiative, lets people request to have their identifying details removed from old stories.

These efforts are important — but Weinman’s concern is that, in the case of a crime in the news, “there is still a story, and it’s often entirely reliant on the word of the police department.” The biggest question for her, she said, “How soon should you even publish that story? And why is it that police sources are more authoritative than community sources? Can’t they be given equal weight? I think these are questions that just have not been adequately resolved.”

Cassie Owens, who’s taken on and expanded the mantle of Chappell and Rispoli’s work around changing criminal justice reporting in Philadelphia, echoes Weinman’s concerns. “One of the things that is tough with local news is that the coverage we all know would be better in terms of its informational value — and that we all know is possible — takes more resources to produce than crime briefs,” she tells Nieman Lab deputy editor Sarah Scire, in an interview running today.

Local news staff writer Sophie Culpepper looks at the efforts of three local digital news organizations — The Triibe, Outlier Media, and The Berkeley Scanner — to cover crime differently, and finds that audiences are sometimes responding in surprising ways.

“A whodunnit with cliffhangers and a strong narrative thread.” Serial spawned worthy successors and flimsy imitations, TV shows, and novels. It faced plenty of backlash, was acquired by The New York Times (“The idea here is to have more Serials”), and released Season 4, “Guantánamo,” earlier this year. True crime remains a fixture of Apple’s podcast charts, where it has its own category. Staff writer Neel Dhanesha uses the genre as a lens to look at how public radio has changed over the past decade — and finds that public radio executives see their success with true crime as a model for all types of investigational podcasts going forward.

“I have seen pages where people do their makeup while talking about true crime.” In the first episode of Serial, Sarah Koenig stressed that she was “not a detective or a private investigator. I’m not even a crime reporter.” Koenig was, however, a long-time journalist. Today, many of the people covering true crime are not. Staff writer Hanaa’ Tameez dives into the world of TikTok true crime, speaking with four creators who aim to produce “ethical” true crime content, with empathy for victims and their loved ones, while also making money. “I think the reason why [families] reach out to either me or to other creators,” one creator tells Tameez, “because they know it’s way quicker to get a story out through me than through Univision and Telemundo.”

“It is extremely prohibitive, just in a really simple mechanical sense, if you wanted to say, how does this compare to all cases?” In 2022, Cleveland.com received $200,000 from the Google News Initiative to systematically unpublish thousands of stories that would qualify for right-to-be-forgotten requests and thousands of mugshots. “An editor’s eyes were on everything, but the automation helped bubble up the material,” Cleveland.com editor Chris Quinn told me recently. Today, nearly all of the mugshots published on Cleveland.com in the past 10 years have been removed from its website.

The Cleveland.com initiative was a relatively early example of machine learning’s potential in reshaping crime reporting at a large scale. But more complicated projects involving large language models are already in the works. AI staff writer Andrew Deck looks at promising efforts from The Marshall Project’s “Testify” investigation in Cleveland, Ohio, which show the power of computational reporting on county courthouses. But Deck finds, too, that this type of reporting is difficult and expensive to replicate.

1. One note: Throughout this package, we refer to both “crime” and “true crime.” From what I can tell, “crime” and “true crime” are useful terms to distinguish types of books from each other — “crime” fiction from “true crime” nonfiction — but less useful when it comes to journalism, the focus of this package. (Technically, all crime journalism should be “true crime” because it should report things as they actually happened.) In news, a more useful distinction between “crime” and “true crime” might be that “crime” encompasses broad trends and statistics while “true crime” refers to more individual stories with plots and cliffhangers. [↩]