Last month, WBEZ – the folks who bring you “Wait, wait don’t tell me!” – sent out 77 new newsletters on the same day. That’s one for every neighborhood in Chicago. Each newsletter featured unique statistics and charts about that specific neighborhood – stories on the number of traffic collisions, tallies of complaints about rat infestations, even data about recycling.
But WBEZ staff members didn’t write 77 different newsletters. They wrote one. Software from Crosstown automatically inserted the charts, graphs and stats for each neighborhood, allowing WBEZ to turn the work of one reporter into stories tailored for dozens of communities.
“It’s exciting to offer content that’s tailor-made for members of our audience,” said Alden Loury, the data projects editor at WBEZ.
In North Carolina, WRAL authored one newsletter, but created 18 different versions for all the neighborhoods in Raleigh.
The initiative is inspired by the current rallying cry for local news: “Build your own infrastructure.”
For the past two decades, local publishers have eked out an existence on the backs of platforms built by Big Tech. Anyone who winces at the term “pivot to video” knows what I’m talking about.
Newsrooms twisted themselves like a pretzel to fit into this architecture. Journalists wrote headlines and ledes not for their audience, but to please SEO algorithms. Editors chose photos based on how they performed on Facebook.
There are plenty of reasons why publishers became Silicon Valley client states. Big Tech employs the best engineering talent, and the companies have the resources to build platforms with an ever-expanding array of features that are easy to use and rarely break. No wonder they have all the money.
But they change their focus like a teenager changes outfits. In fact, the experiment Crosstown is running with WBEZ and WRAL was launched with funding from a Meta Journalism Project grant managed by Local Media Association. Meta has since ended its journalism philanthropy arm and Crosstown almost ran out of money.
It’s clear that journalism, particularly at the local level, needs its own technical infrastructure built specifically to serve its own needs. Crosstown grew out of a partnership between the University of Southern California’s journalism and engineering schools. Only by teaming up with bleeding-edge experts in computer science were we able to build a product that was able to give local news something it had been sorely lacking: scale.
The Crosstown newsletter platform allows a publisher to produce different versions of the same story that are then fitted with the data and charts for every community. A piece about burglaries in Los Angeles, for example, becomes 114 different stories, each with the burglary rate, percent change and rank for a particular neighborhood.
One month into this experiment, WBEZ is finding it can now efficiently produce journalism for parts of the city that have been overlooked in the past. “We’re not only providing them data, visualizations and context about what’s happening in Chicago,” said WBEZ’s Loury, “but we’re also able to give them a slice of that broader story that’s happening in their very neighborhoods.” Open rates on its newsletters have been above 70%.
We’ve been working on this project for some time. We’ve struggled with parts of it that we thought might be simple. There is still a lot we need to improve. But it makes sense to build one tool that can be used by many.
If each newsroom tried to collect all its own data on crime, traffic or housing, the cost would be enormous. If Crosstown can build one tool that numerous newsrooms can use, the efficiency benefits all.
Still, building good software is neither cheap nor easy. The launch of Press Forward, the $500 million philanthropic initiative to support local news, should inject some much-needed R&D into this space. In fact, Crosstown won’t be able to reach its goal of serving 100-plus newsrooms without significant investment in its technology. But, as our pilot projects with WRAL and WBEZ are demonstrating, the potential payoff is enormous.
Gabriel Kahn is professor of professional practice at the USC Annenberg School of Journalism and leads the Crosstown data journalism project.
The LMA/Crosstown Data Journalism Pilot initially launched in 2021 in partnership with Local Media Association with funding from Meta Journalism Project, supporting data journalism at NOLA.com/The Advocate, WRAL and WBEZ. Read previous case studies: