Campaign Strategy Research Reveals Unexpected Risk in Base-Mobilization Messaging
Political campaigns may be wasting millions on a communications strategy that backfires more often than it succeeds, according to new research from the Wharton School that carries implications for how corporate leaders think about stakeholder messaging during high-stakes moments.
The study, led by Wharton marketing professor Pinar Yildirim, found that campaign speeches designed to energize a candidate's core supporters—what political operatives call "firing up the base"—can actually hurt electoral prospects depending on how the media covers them. The findings, published February 17, suggest that the massive sums campaigns now spend on communications may be poorly allocated, with messaging strategies built on assumptions about voter behavior that don't hold up under scrutiny.
The research arrives as corporate finance leaders increasingly face their own version of the same dilemma: how to communicate during crises, activist campaigns, or proxy fights without alienating the stakeholders they're trying to mobilize. The parallel isn't perfect, but the underlying question is identical—when does rallying your supporters create more problems than it solves?
Yildirim's team examined how different types of campaign messages affect voter behavior, focusing specifically on the interaction between message content and media coverage. The study's core finding challenges conventional political wisdom: base-mobilization speeches that generate significant media attention can shift outcomes in unexpected ways, though the research summary provided does not specify the direction or magnitude of these effects.
The study comes at a moment when political campaigns are raising "vast sums of money," as the research notes, with much of that capital flowing into communications infrastructure. Yet despite this spending, the researchers found it "remains unclear" which message types actually change voter behavior—a knowledge gap that has persisted even as campaign budgets have ballooned.
For corporate finance executives, the research offers an uncomfortable mirror. Companies routinely deploy communications strategies during shareholder battles, regulatory challenges, or reputation crises with limited evidence about what actually works. The assumption that energizing your core supporters is always beneficial—whether those supporters are retail shareholders, employee-owners, or key institutional investors—may be as flawed in corporate contexts as Yildirim's research suggests it is in political ones.
The mechanism appears to hinge on media dynamics. A message crafted for internal consumption (rallying the base) can take on different meaning when amplified through news coverage to broader audiences. What plays well in a room full of true believers may alienate the persuadable middle when it reaches them through journalistic framing.
The research doesn't provide a simple prescription for campaign managers or corporate communications teams. The interaction between message type and media coverage creates a complex decision tree where the "right" strategy depends on predicting not just how your core audience will respond, but how that response will be covered and how the coverage will land with everyone else.
What the study does make clear is that the current approach—spending heavily on base-mobilization messaging without clear evidence of its effectiveness—represents a significant misallocation of resources. Whether those resources are campaign dollars or corporate communications budgets, the underlying waste is the same: money spent on strategies that haven't been proven to work and may actively backfire.
The findings suggest a need for more rigorous testing of communications strategies before deployment, particularly when the stakes are high and the audiences are fragmented. For CFOs overseeing crisis communications or investor relations budgets, that's a framework worth importing from the political arena—even if the specific tactics need translation.


















Responses (0 )