innovations

‘Structured journalism’ offers readers a different kind of story experience

July 30, 2015

Sign up for The Media Today, CJR’s daily newsletter.

It happens thousands of times a day. A journalist gets an assignment, reports out the story, over days or even months, and then sits down to the process of writing–struggling over the lede, debating how to structure the piece, and then crafting the most impactful, poignant story possible.

Once published, the story might be accessible on various platforms and devices, but there is only one way to interact with the finished product. One size fits all; no moving of the pieces for readers who want a different experience.

But what if the constraints of the written article were dissolved? What if a story could be personalized to each reader? That aspiration is at the core of what a growing group of journalists hopes to create through a practice called “structured journalism,” an umbrella term that refers to thinking of journalism as bits and pieces of information that can be mixed and matched in infinite ways.

With a “structured” approach, information–the journalist’s currency–is modular and accumulates over time. Readers can enter at the point of a story that’s most relevant to them, then explore the rest at their own pace, guided by their personal curiosites.

This idea, which is gaining hold in newsrooms like The Boston Globe and The Washington Post, challenges the supremacy of the narrative, non-interactive form that, despite an explosion of digital innovation, remains the cherished mainstay of journalism.

“Structured journalism plays so nicely with the affordances of digital media,” says David Smydra, a former Nieman fellow and current editorial manager at Google Play Newsstand. “It not only produces incredible stories but creates this reservoir of material that reporters and readers can call upon for future stories.”

Sign up for CJR’s daily email

A example of this trend is the Knowledge Map, an experimental project from The Washington Post. Its debut article about ISIS’ impact on tech companies is full of annotated text (such as names, themes, and events) that when clicked open summaries, graphics, or other supplementary material in a side panel. At certain points within the story, highlighted questions like “What is ISIS?” appear for readers who need more background. The resulting story reads like a typical article, only with opportunities to click on a side panel for a moment or delve deeper into the paper’s archive.

The hope is to replicate this technique across other stories, says Sarah Sampsel, director of digital strategy at the Post and the project’s designer; looking forward, Sampsel would like to add tools to the newsroom that would allow writers to participate in the process. “We think there’s an opportunity to better connect the journalism we create,” she says.

The Knowledge Map provides one possible solution to the redundancies that often plague the traditional journalistic format, in which background and contextual material is repeated across articles. In the existing model, readers who have been following a story are given too much information, while newcomers are given too little.

The Knowledge Map is one application of the principles underpinning structural journalism, but it’s only one example of what is still a relatively undefined trend.

Circa, a mobile-first app that died earlier this year, is often cited as a prime example of structured journalism, what its creators called “atomized news.” Circa deconstructed the news into units–facts, quotes, news updates–and used these units to update continually developing storylines. Readers followed storylines as they would a Twitter feed, receiving only the newest updates to those overarching narratives.

“When you break the story down into something smaller, you get something bigger,” says David Cohn, former chief content officer at Circa, because all stories are interrelated in an ever-extending web. At Circa, the editorial process involved mapping this web of stories and choosing where each new story component belonged.

Bill Adair, creator of Politifact and one of the early proponents of structured journalism, says one of Circa’s flaws was that it wasn’t structured enough. “Circa was like traditional journalism with a good workout for your thumb,” says Adair. The app deconstructed a story, but still  reassembled it in the linear format of a news article.

Adair and other structured journalism proponents would like to upend traditional journalistic formats. Structured journalism, they say, could evolve into a unified theory of digital journalism–though no one yet knows what that might look like.

The idea presents questions like: What if every storyline had its own URL? What if a newsroom created an in-house journalistic Wikipedia? What if articles could be parsed like code? What if the basic structure of journalism allowed users to explore every story through timelines, social graphs, and maps, instead of requiring special, one-off projects?

Because structured journalism overlaps with computer science, data formatting, and information design, it often–and wrongly–gets confused with data journalism. Smydra explains the difference: “Data journalism is the practice of analyzing data in order to unearth new stories,” he says. “Structured journalism is the practice of turning one’s reporting into data that can be repurposed in any number of ways.”

An outlet that did this well was Homicide Watch DC, which collected data on every murder in Washington, DC, over the four years of its existence. The novelty of Homicide Watch, which closed in 2015, was in the premise that every murder is worthy of the resources required to report it–unlike standard journalistic practice, which bases homicide coverage on a magic formula of public interest, news judgment, and sometimes physical space in a newspaper. (The original Homicide Watch closed after four years, but a partner project in Chicago, and the L.A. Times’ similar Homicide Report, continue.)

The product is a visual database that can be queried, tallied, searched, and tracked. 

The editorial process at Homicide Watch and similar projects centers on what data to include in the collection process, and how to present it.

“When Homicide Watch was set up,” says David Caswell, founder of the structured journalism site Structured Stories, its creators “made decisions that would limit what they could cover and what fields they were going to use.” Homicide Watch collected information on race and age, but not income, for example.

Laura Amico, who co-founded Homicide Watch DC and is now an editor at The Boston Globe, applied the same principles to the Globe’s coverage the federal trial of Dzhokhar Tsarnaev, one of the two brothers responsible for the 2013 bombing of the Boston Marathon. But she faced a conundrum. Homicide stories usually take a fairly simple narrative structure, which repeats itself, but a trial would be a long-running story with narrative twists and turns. So Amico and her team tried to think of creative ways to structure coverage of the trial.

What they came up with was a witness and evidence network, which published daily updates demonstrating how new pieces of evidence and testimonies related to each other. The network was one way readers could explore the trial, along with a day-by-day timeline and traditional coverage.

“People could navigate through all the coverage, no matter which story they landed on,” says Anica Butler, assistant metro editor at the Globe. Still, in hindsight, Butler says she’s not convinced the labor-intensive upkeep of the project was the best use of resources. “You have a ton of space online. You can put anything online, you can put everything online.” The real question, she says, is “will it help the reader to understand the trial?”

Reginald Chua, an executive editor at Reuters and an early developer of structured journalism, says that structured journalism doesn’t only offer better ways to tell certain stories, it introduce new forms of journalism not considered before, which can create tension in newsrooms. Aside from the practical challenges of newsroom adoption, the question of what gets lost in structured journalism will need to be addressed. “Frankly, the narrative outcome [of structured stories] is not as exciting,” says Chua, “That was one of the big complaints over the death of Circa.”

But the investment can lead to an entirely new–and possibly more successful–way to cover complex, long-running stories. “In a structured approach, the journalism accumulates,” says Caswell, “A single journalistic artifact could accumulate over months or years, in an organized way.”

“I think it’s a fallacy to think that reporters haven’t been doing structured journalism all along,” says Smydra. By collecting information and choosing what matters, “they’ve just been skipping the step of annotating it in a structured way.”

Chava Gourarie is a freelance writer based in New York and a former CJR Delacorte Fellow. Follow her on Twitter at @ChavaRisa