A number of news organizations have built their own analytics dashboards. While we won’t be able to go over each of them in depth, the links below (and end citations) provide further detail.
Although our primary focus in this report is on investigative newsrooms, other organizations with different goals are also experimenting in this space. Carebot is the NPR Visuals team’s project to capture how their projects, often human-interest and photography-based, affect their audience. “What does impact for a team like ours look like?” asked Brian Boyer, editor of the Visuals team. “We came to the realization that what we create is empathy—we try and make people care about someone else. Carebot is finding ways to measure if we made people care or not.”
An open source project, Carebot focuses on user actions44how many liked it on Facebook) but with an added twist: It divides that number by the total unique visitors for a given story. This metric allows the team to say things like, “thirty percent of all people that read this shared it in some way.” Such statements let them more easily compare articles while controlling for variations in page views or total traffic.
The other part of the project involves adding questions at the end of a story, such as, “did you love this story?” or “did you learn something from this story?” If users answer “yes,” they are led to like the story on Facebook or donate to the station. These questions aim to bring people into the public radio family and are the result of thinking about user experience and user flow as a crucial part of the impact question. (Example: After a reader finishes a story, what should he or she do? And if we can think of actions we prefer over others, how can we optimize and measure that?)
“There are stories that are going to have great raw numbers because they are about a celebrity comedian that’s going to host The Daily Show and the controversy about his tweets: that’s just going to succeed,” Boyer said. “So how do you take work that is more to the mission of the organization and hold it up to say, ‘this thing might not have the page views, but it’s doing the mission.’ Carebot is about ‘how do we prove we’re doing the mission?’ ” Indeed, mission-driven metrics is a good label for this type of thinking and, as we’ll discuss later, highlights the crucial intersection between successful impact measurement and stated organizational goals.
Also working at NPR, Melody Kramer and Wright Bryan designed and built an internal dashboard based specifically on the questions their editors and reporters had in the course of a news day.45further, their user-centered design led them to frame visualizations in a friendly and inviting way and served as a great source of inspiration for parts of NewsLynx. Their platform took shape after hours of interviews with staff, focusing specifically on their daily decisions and how technology could help them arrive at smarter decisions faster.
ProPublica is another outlet taking significant strides to measure its impact. In a 2013 report, Richard Tofel, ProPublica’s president, outlined the organization’s approach to tracking impact:
ProPublica makes use of multiple internal and external reports in charting possible impact. The most significant of these is an internal document called the Tracking Report, which is updated daily …The report records each story published …and any prominent reprints or pieces following the work by others (with most of this data derived from Google Alerts and clipping services). Beyond this, the Tracking Report also includes each instance of official actions influenced by the story (such as statements by public officials or agencies or the announcement of some sort of non-public policy review), opportunities for change (such as legislative hearings, an administrative study or the appointment of a commission) and, ultimately, change that has resulted. These last entries are the crux of the effort. They are recorded only when ProPublica management believes, usually from the public record, that reasonable people would be satisfied that a clear causal link exists between ProPublica’s reporting and the opportunity for change or impact itself.46 Tofel goes on to explain that ProPublica tracks these outcomes for years after an article’s publication—“possible prosecutions and fines continue to result from this work long after the reporters involved have moved on to other work, and ProPublica notes these as they emerge.” In tracking the impact of its work, ProPublica has also developed sophisticated tools like PixelPing,47allows for measuring traffic to its articles that have been republished on other sites.
The Guardian uses a custom system called Ophan that is tuned to the needs of editors so that they can quickly see what is being read on the site and start to explain the why behind those trends.48linked walkthrough, however, is a good starting point.
One typical newsroom-specific problem is how to understand the way readers engage with a package of stories. Editors have ideas about how readers should navigate between these pages, but few teams are tracking how they actually do so. To better understand these flows, James Robinson built Package Mapper to track the multi-story experience.49miles away from the simplistic page view, this project looks at user experience and user flow as a key part of understanding performance.