That's the question Sudha Ram, from the University of Arizona, set out to answer in a recent study of a dozen major news organisations that use the social media website for sharing their content.
The answer, according to Ram's research, varies widely by news agency and there may not be one universally applicable strategy for maximising Twitter effectiveness.
However, news agencies can learn a lot by looking at how their news diffuses once it is posted on Twitter, said Ram, professor of management information systems (MIS) and computer science at Arizona's Eller College of Management.
Ram examined, over a six-month period, the Twitter activity of 12 major news organisations focused on US news, global news, technology news or financial news, according to an Arizona statement.
All of the agencies selected - the New York Times, Washington Post, BBC, NPR, Reuters, Guardian, Forbes, Financial Times, Mashable, Arstechnica, Wired and Bloomberg - regularly share news articles on Twitter, which allows users to post 140-character messages as well as links to online content.Ram, working with Devi Bhattacharya, an MIS doctoral student at Arizona, tracked what happened to a news article after it was tweeted by a news organisation.
Together, they looked at how many people retweeted, or reposted, the article on their own Twitter feeds, then how many times it was subsequently retweeted from those accounts and so forth.
They were then able to evaluate the volume and extend of spread of an article on Twitter, as well as its overall lifespan.
"The goal for a news agency is to have a lot of people reading and following your articles," said Ram.
"What we've done is use network analysis, which is quite different from just looking at the total number of tweets or total number of retweets. You're starting to see, over time, how information is spreading," Ram said.
Ram and Bhattacharya rendered the data they collected from each organisation visually as images. These appear something like fireworks, with dots representing individual Twitter users and cascade streams from those dots depicting retweets.