More and more organizations are turning to social media to reach as many potential customers as possible, yet few marketing campaigns truly “go viral.” This is largely due to the immense amount of content competing for attention in social networks. Even if users see a particular campaign, they may pay no attention to it, due to the overabundance of information in their feeds. In a timely international collaboration, Ralf van der Lans of HKUST and two co-researchers propose a novel strategy for maximizing campaign reach by accounting for competition for attention on social media.

“To initiate a campaign that reaches many people,” say the researchers, “a firm first needs to define a seeding strategy.” This involves identifying a small number of key individuals capable of propagating the firm’s marketing message as widely as possible, via a “cascade” effect. Such “seeds” should be “well-connected network members,” say the researchers, “who are able to reach many individuals quickly.” However, another criterion must be considered. “Effective seeds should not only have many friends,” the authors point out, “but their friends should also be susceptible to incoming information.”

Drawing on exchange-network theory, the authors proposed an innovative seeding strategy: targeting “social network members who have many friends but whose friends have only a few friends.” Such members are the most effective seeds, the researchers hypothesized, because there is little competition for their friends’ attention, making their friends more likely to pass on the campaign content.

While many studies have looked at competition for attention on social media, they have failed to consider its implications for seeding strategies. The researchers are the first to theoretically derive and empirically validate an optimal seeding strategy focusing on competition for attention. The team tested the proposed strategy in two studies involving 34 real-life social media campaigns and found that their novel approach could indeed substantially increase a campaign’s reach.

In their first empirical study, the researchers analyzed the spread of a viral game-based campaign launched on a large social media platform to promote an animated movie. The results indicated that “the proposed optimal seeding strategy, which accounts for competition for attention, outperforms benchmark seeding strategies by up to 70%.” To test the generalizability of their strategy, the researchers conducted a second study of 33 campaigns during the 2010 Super Bowl period, when many brands launched new advertising initiatives. They explored how the campaign messages competed for attention within a social network of undergraduates at a major US university.

Supporting the researchers’ theoretical predictions, the results of both empirical studies suggested that firms can maximize the reach of their social media marketing campaigns by selectively targeting users—specifically, network members who are themselves well connected (often known as “influencers”) but whose followers do not have many friends. The researchers found that this approach was up to three times more successful than targeting network members at random. “Our results hold across many campaigns involving different content, different sharing processes, and different platforms,” they add.

These findings shed new light on how information is disseminated on social media, showing for the first time that “competition for attention is the underlying mechanism.” In an era of ubiquitous social media use, this has critical implications for managers and marketers seeking the optimal seeding strategy to increase customer engagement online.