How do visuals in company crisis messages shape perceptions of a crisis?

Many of us are familiar with iconic photos following environmental crises, such as birds covered in oil after the BP oil spill or the devastation of recent wildfires in Hawaii. Visuals can catch our attention and shape our perceptions of events.

In this study, we wanted to explore how visuals used by companies in crisis response messages might impact stakeholders’ perceptions of the crisis. For example, does using different images in a response message lead to different emotional reactions and beliefs about responsibility for the crisis? Can visuals impact people’s perception of the company’s reputation and their likely behavioral actions toward the company?

We conducted an online experiment to answer these questions. We drew on existing theories, like exemplification theory, to decide which kind of visuals to test. Specifically, exemplification theory suggests that examples can make messages more memorable and persuasive. However, examples can take many forms. Accordingly, we crafted corporate social media responses to a crisis showing a picture of the CEO speaking out about the event, a group from the company engaged in repair work, an infographic about the company response, and a text-only message.

We designed the company response as a threaded social media post (like on Twitter/X) to mimic the kind of public outreach a company would do in a real crisis. For more generalizability to real-world crises, we used two crisis types: a food recall and a chemical spill. We also designed the posts as coming from two real companies, General Mills and Kraft Heinz (and we let participants know at the end that this was a fictional crisis).

So what did we find? Did the type of visual accompanying the company’s response message shape people’s perceptions of the crisis?

In short, no. There were no statistically significant differences on any of our attitude or behavioral variables based on the type of visual participants saw. In other words, the type of visual (e.g., CEO photo, group employee photo, infographic, or text-only message) did not lead to different levels of sadness, anger, fear, attributions of responsibility, transparency perceptions, reputation, negative word-of-mouth intentions, or positive company perceptions.

However, we caution readers not to interpret this to mean that visuals don’t matter in crisis situations. Rather, it could be that people paid more attention to the textual element of the message, so the visual wasn’t as impactful for them in this study context.

Another goal of this project was to consider the ethics of visual selection in a crisis, and we offer that as a final takeaway for readers. We drew on an ethical model of persuasion that encourages practitioners to think in part about the authenticity, truthfulness, and respect communicated to readers. Visuals shared in crisis situations should also be rooted in ethical intent and not considered only as an afterthought. Particularly with social media being heavily visual, we encourage everyone to think carefully about ethical visual selection and ethical data visualizations.

For instance, make sure that data visualizations present accurate, complete information, which involves transparency, and that photos showing company employees are actually engaged in the repair work (not just putting on a photoshoot). In short, photos should match the off-camera reality of the company’s response.

In future research, we may explore other emotional responses to more vivid visuals (such as threat-awe in a natural disaster) and further parse out ethical crisis visual strategies, such as gauging stakeholders’ perceptions of authenticity or social responsibility from images. For example, we learned from our pilot test results that some participants did not like the image of the CEO smiling in the crisis response message. This kind of image displaying an emotional reaction could be perceived as inauthentic or disingenuous to readers, and examining emotional reaction is worthy of future visual crisis communication research.