A field guide for assessing chances of online-to-offline mobilization 

9 June 2023 

By Jared Holt and Katherine Keneally 


In March 2023, media, elected officials and law enforcement were expressing fear that supporters of Donald Trump would engage in violent, large-scale protests after a New York grand jury indicted the former president, with many devoting significant resources to answer the perceived threat.  

But observations of online chatter before and after Trump’s indictment had shown few, if any, indications that a mass mobilization of Trump supporters was imminent. Rather, observations of online communities across the political spectrum with proven capabilities to produce waves of offline activity suggested that threats might instead spawn from individuals or small groups seeking to intimidate figures involved in the indictment. Although few people attended post-indictment protests, some individuals menaced prosecutors online and offline—an outcome reflecting the online ecosystems such activity often emerges from. 

Potential protests associated with increasingly polarized political causes continue to be subjects of mass-media hype cycles, influencing and engendering hysteria around the event. This panic is often driven by non-expert pundits and sensationalist media outlets who lift hand-picked comments from the internet — such as talks of civil war or fantasies of executing politicians – and strip them of crucial context, leaving their audiences with incomplete pictures of a threat environment. Though dehumanizing and violent rhetoric online is concerning, the presence of such language is not always a promise of the offline havoc it is thought to precede. 

This is not to say that vigilance surrounding a high-profile political event is not important, but that there is a difference between alertness and panic. With proper risk analysis of online environments surrounding an event or incident, researchers, media, and law enforcement alike can better assess whether large-scale mobilization is likely to occur, better informing their decision-making and abilities to anticipate and deter potential violence. 

ISD analysts understand the repetition of this cycle as a byproduct of incomplete analysis. To address this, we have provided a list of key factors for consideration when assessing turbulent online information environments (forums, groups, chats, regular posts, videos, etc.) for their likelihood of producing large offline actions. Those who track online threats after major events should consider the following when contextualizing the rhetoric they encounter and whether it indicates a potential for larger offline actions or incidents of violence. 

Factors for Consideration 

  1. What communities are most likely to react? This requires identifying the specific communities that are most likely to react to an event and who may engage in offline activities in response. While individuals may respond by posting violent rhetoric in the comments section of an article, these standalone posts provide little information on whether large-scale protests are likely to materialize. Instead, researchers should analyze the posts amongst communities identified as most likely to respond through offline activities to determine whether groups are discussing mobilization. Additionally, one should also look to the group or community’s history of previous activities, such as participation in previous protests or other forms of offline activities (e.g. petitions), to assess their likelihood of mobilizing.  Background knowledge and community familiarity are essential to this assessment process.
  2. How important is the event to the targeted communities? Analysts should evaluate the context of the development or event and determine the importance within the group in question. Rhetoric surrounding the 2020 election indicated that many participants in the January 6th Capitol riot believed that the election was stolen and the country needed to be saved. Responding to the 2020 election via mass mobilization was understood by some as imperative to the future of the United States. Large-scale political mobilization becomes more plausible when events or developments are viewed as an existential threat to these communities. Many major events or developments related to a common cause (e.g. abortion rights) can shape conversations within communities online or integrate into existing narratives (for example, ‘groomer’ slurs being used in posts responding to news relating to drag performances). However, many fail to motivate individuals to engage in large-scale offline efforts. Smaller-scale actions (like anti-LGBTQ+ protests at drag shows) can still be more likely, as they require less planning and fewer people to execute.
  3. How specific are the calls for action? Vague mentions of “protest” do not provide significant insights into whether a protest is likely to occur, let alone a violent protest. Potential protest participants require specificity in the call to action to know where to go and when. Historically, successful calls for mass action have offered would-be participants a clear vision of the action and desired outcomes. Locations and times for actions provide a timeframe and deadlines, and clear goals offer direction and motivation. The presence of conversations related to logistical details involved in engaging in an offline action, such as the sharing of travel plans and supporting event infrastructure, can indicate that the proposed action is more likely to happen.  
  4. Is there support for mass mobilization from high-profile, trusted figures? Calls to action gain legitimacy when those considered trusted community leaders help build support, whether directly or indirectly. These messengers—including politicians, media figures and social media personalities—offer perceived legitimacy to the cause and can amplify the call to action to wider audiences online. In some online spaces where users largely operate under pseudonyms, community consensus can offer a substitute for more singular voices, which can then be supported by these figures. 
  5. How do these online communities perceive violent rhetoric? The presence of violent or extreme rhetoric can give cause for concern, particularly when it appears alongside other elements on this list. When considering the impact of that rhetoric on a larger online community that may produce large offline actions, one must consider that community’s broader reaction to violent content. A lone comment on a thread that receives little engagement is often of less concern than such comments appearing repeatedly and receiving encouragement from others in a community.
  6. Do the calls for action have significant reach across platforms? The repeated appearance of motivating sentiments, violent rhetoric, leadership buy-in, and calls to action across several platforms and online communities can offer clues as to the scale of potential participation in a would-be mobilization. When observed, repetition can also help to reinforce or question earlier analysis. Wider consensus across platforms and communities can offer a larger base of individuals who may be persuaded to engage in offline action.