RJI Fellow developing tool to measure real-life impact beyond clicks and social shares

Fellow believes showing a story’s impact can help in the process of restoring trust

The Donald W. Reynolds Journalism Institute awarded seven fellowships for the 2019-20 academic year with projects to improve gun violence reporting, expand solutions-based journalism by local TV stations, help large and small newsrooms get the most out of push notifications, customize audio documentaries through voice commands, measure the community impact of online stories and preserve digital content that’s being lost.

Finding the true impact of a story should go beyond just looking at clicks and social shares, says Leezel Tanglao, a data-driven multimedia news leader. Although these data points are important, she adds, tracking impact should include finding out if a story made a difference in someone’s life. Did the story prompt proposed legislation, affect a company’s stocks or inspire a conversation about a community issue?

Unfortunately, tracking this real-life impact can be harder to do, she says, without searching the internet manually for possible connections. That has challenged her to develop a tool during her fellowship that provides an objective, actionable and holistic view into how a news story affects the community.

“Impact must be seen … in a more holistic way through the lens of a diversity of views, the implication for underserved communities and business models,” she says.

Her tool will also uncover opportunities for potential story follow-ups.

Showing real impact from news stories could help restore public trust and dispel fake news, she says.

“When you are able to show impact in a real and tangible way, you can continue to build trust with readers,” says Tanglao. “And news organizations are able to fulfill their missions.”

How the proposed tool works

Tanglao’s tool, which is in development with the help of UCLA’s engineering school, will tap into multiple open-source databases and APIs using artificial intelligence and natural language processing.

Journalists will run a story URL through the tool. From there the tool would look at key phrases within the story and automatically categorize types of impact this story could have, whether that is an economic, social or policy impact. With this information, it will then ping open-source APIs and databases to find related reference materials, research and other articles.

Tanglao says the tool will dig deeper than an internet keyword search where the content you retrieve may or may not be related to what you are looking for. Instead, her tool’s natural language processing allows you to look for relationships between words and phrases in the article.

“Relationships are more powerful than words,” says Tanglao. “By looking at relationships, we can better surface up relevant and related content.”

For instance, a city council passes a bill. Ideally, the tool would surface any content that was related to that passed bill afterward, she says. Perhaps it prompted an investigation, inspired the formation of a community group or prompted people to submit a public comment.

Tanglao says there are similar impact trackers in the marketplace, but it is her desire to build on what’s already been developed and dig even deeper to develop an automated tool that finds the relational connections. 

Developing a tool like this does pose some challenges, she says. One challenge is figuring out how to teach a computer to detect a relationship between two different things that are relevant and related to a written story.

What if the writer uses “bird” when referring to a plane? Tanglao says the computer would need to be taught to recognize that in this case, the writer is referring to a plane rather than an animal when it says “bird,” she says.

“It’s like you’re basically building almost a human brain in a sense,” she says. “So it’s a challenge because as much as technology has come so far, there’s still a long way to go to really get it close to the way we think as humans.”

How this project came about

Tanglao says she’s been thinking about the subject of impact for a number of years, but more recently since working at The Associated Press and as a consultant to nonprofit businesses as part of her business, StatFury LLC.

Although she didn’t work directly with tracking impact at the AP, she says she saw a gap when it came to understanding the full impact of stories at the organization. According to Tanglao, part of this came from the challenge of being able to track stories adequately since the wire service syndicates its content and customers receive it in various ways.

In addition, she says while working as a consultant, she did some grant reporting writing for nonprofits. The reports required granular details on how funding was making an impact, she says.

“I realized just simply listing off vanity metrics was doing a disservice to the full impact of efforts,” she says. “So with those two things concurrently happening at the same time, it clicked in my head, ‘why aren’t we tracking impact alongside vanity metrics?’”

She began thinking about developing a tool even before she began working at the AP, but she says she didn’t have the bandwidth to take on such a project.

More about Tanglao’s background

Tanglao recently left the AP and is now leading membership and innovation initiatives at HuffPost.

She considers herself a “bridge journalist,” which means she has worked in newsrooms reporting stories for more than 15 years, yet she understands the product, business development and marketing side of the industry.

She helped develop Social Surge, a social metric tool for CNNMoney, with a small team and no previous experience launching a product. Having worked in a variety of newsrooms — including a newspaper, CBS local station and wire service — gives her a more holistic view of the industry, she says. It will also allow her to leverage her diverse network for advice, user testing and connections, she says.

If you're interested in testing Tanglao's tool and receiving updates about it, sign up on this Google document

Related Stories

Expand All Collapse All
Comments

Comments are closed.