Like all Nieman Lab stories, the one you’re reading right now will be anchored by a handful of tags connected to the themes, people, and companies mentioned within. Theoretically, keywords like “tags,” “tagging,” “article tags,” and “metadata” would be helpful in directing people to this, or other, stories, about the mechanics of news discovery.

But do all publishers use article tags the same way we do? And do tags make any difference in attracting people to a website? A new report from analytics company Parse.ly found no statistical link between the use of tags and the amount of traffic a site received — but also saw publishers using tags in creative ways to help guide strategy. (Sites that use Parse.ly include The Atlantic, Slate, Mashable, Business Insider, The Globe and Mail, The New Republic, and Upworthy, among about 400 total.)

In that network, 70 percent reported using tags and 30 percent said they go without. On average, 450 tags were published in a month, with the actual number of tags per post averaging 5.2. Parse.ly found no correlation between the number of overall tags published and audience size, and the number of tags per story didn’t change much with story length. ” Essentially, if we increase the word count of an article from one word to 10,000 words, the average number of tags in the article only increases by around 0.36,” the report found.

(Here at Nieman Lab, we’re relatively aggressive taggers, with over 13,000 tags used on our roughly 5,000 articles published since 2008. But our tag pages account for only about 1 percent of our pageviews. Our 10 most common tags? The New York Times, Twitter, business models, Facebook, advertising, social media, Google, paywalls, the Knight Foundation, and mobile.)

Despite the fact that tags are usually user-facing, their most important uses may be internal to news organizations. “In the last couple of years, we’ve noticed that tags are being used in more interesting ways,” said Andrew Montalenti, chief technical officer at Parse.ly.

Parse.ly’s findings suggest that publishers are making advances in how they use structured data in content production and audience analysis. Companies are moving beyond simply using tags as a way to surface stories on Google or in an archive. Specifically, Parse.ly found that publishers are using tags to categorize article formats (stories vs. quizzes vs. videos), paywall status, and sponsored content.

That same organizational ability to find, filter, and sort stories is being used to route different processes in the newsroom, Montalenti said. A site that runs native ads could track and report on specific campaigns based on tags for native content and the accompanying brand. Or, in the case of a site like The Globe and Mail, tags could be used to indicate which stories are only available to subscribers and which are available to everyone.

Last year, Joshua Lasky, a marketing manager at Atlantic Media Strategies, wrote about new ways how media outlets could use tags:

There’s a certain amount of content optimization that can be done here. Think if you were to tag articles based on whether they included a video; you could then analyze whether having this element improved article performance. This would help you to decide whether it made sense to embed more videos in the future.

In July, The New York Times offered a glimpse of a new automated tagging tool under development in its Research and Development Lab. Alexis Lloyd, creative director of the Lab, said that such tools will let the Times mine its archive to create new products in the future.