Nuzzel unveils NuzzelRank, which scores news sources on ‘authority’

Everyone from Elon Musk to AdBlock Plus wants to tell you which news sources are worth trusting. Now news aggregator Nuzzel is joining in.

Specifically, it’s launching NuzzelRank, which founder and CEO Jonathan Abrams described as “our new authority ranking of thousands of top news sources, using signals from top business influencers.” He said it replaces a more “simplistic” ranking system that it was using for its news monitoring and research product Nuzzel Media Intelligence.

You may also see NuzzelRank outside the company’s Media Intelligence reports. For one thing, there’s a new page with rankings of Nuzzel’s top sources. For another, Abrams said publishers will be able to add badges with their NuzzelRank scores to their websites, and he also plans to make this data available through an API.

At this point, you’re probably wondering how Nuzzel does this ranking. You’re definitely wondering that if you looked at the top sources ranking and saw that TechCrunch is comes in at number four overall. That’s right: We score below The New York Times and The Washington Post, but above The New Yorker and Wired — which is both flattering and a little nuts.

Abrams said there are three main ways that Nuzzel calculates the score. First, there’s data within Nuzzel itself, including the reading behavior of its users. Second, it’s looking at “external signals about the engagement and authority of news sources.”

Third, it’s working with a whole bunch o outside organizations that have developed different approaches to scoring news sources and sorting out which ones are and aren’t trustworthy — so Nuzzel is joining the Trust Project and the Credibility Coalition, and it’s also partnering with NewsGuard and Deepnews.ai.

In the announcement, Abrams emphasized that the company isn’t relying on human editors or making these judgments on its own: “Nuzzel has always focused on building scalable solutions that use software to aggregate existing valuable signals to provide useful results, rather than human approaches that are not scalable and subject to bias.”