Research Explores How to Maintain Trust When Pivoting From Failing Projects

Charles Dorison, assistant professor of management, has been exploring the difficulties of navigating the fine line between abandoning failing projects and maintaining trust from key stakeholders. 

Here, he offers the findings from his co-authored research for how managers and organizations can manage this tension.

What inspired you to pursue this research topic?

This research program started in a collaboration with Drs. Chris Umphres (United States Air Force) and Jennifer Lerner (Harvard University). In our discussions with accomplished executive education participants, we heard over-and-over that abandoning commitments creates political costs in ways our economic models failed to capture. Our first publication (Dorison, Umphres, & Lerner, 2021, Journal of Experimental Psychology: General) provided data to support this argument in the context of de-escalation of commitment to failing courses of action.

But this left me with a puzzle: how can leaders overcome the bind between abandoning failing projects and maintaining trust from core constituencies? Enter Ariella Kristal: an expert in precommitment and a postdoctoral fellow at Columbia Business School. In our recent publication (Kristal & Dorison, 2024, Journal of Applied Psychology), we found precommitment (i.e., a public pledge to change course conditional on a concrete future state of the world) can solve this puzzle. 

What is one surprising or unexpected finding from your research?

We set out on this project to find a communication strategy that would allow leaders to maintain trust when de-escalating commitment from failing courses of action. We hit two major surprises along the way. 

The first surprise was what didn’t work: simply trying to teach people about sunk cost bias. When we started the project, we thought it would be simple: just explain to people why de-escalating is the correct choice. But, this communication strategy was surprisingly ineffective.

The second surprise came a bit later. Once we began studying precommitment, we thought we had found a straightforward and successful communication strategy for leaders to use. But of course, the answer is never that simple. In conducting our studies, our work also identified an important and unexpected—to us—cost of using precommitment. More concretely, we found that precommitment can yield a negative externality: undermining perceived confidence and motivation among followers at a project’s inception.

What are some practical implications of your work that professionals or businesses can use?

We think our work has a very straightforward practical implication. Regardless of profession, any manager will face a point where they need to de-escalate commitment to a failing course of action. As part of that de-escalation, they’ll need to justify their choice to their fellow employees—employees who may be deeply invested in the project. We believe precommitment, when implemented correctly, can be a very useful tool in this context. For example, imagine you are in charge of designing a new marketing campaign. You might precommit that, if the ad sales do not reach a certain level after three months, you will pivot to a new marketing strategy. If after three months the benchmark has not been met, this precommitment can allow you to maintain trust from your team when moving in the new direction. The precommitment does not have to be binding; however, having it in writing can help you avoid investing further resources into a failing course of action. 

What are the next steps for your research? Any upcoming projects or new directions?

Everyone can agree that people are irrational. Together with colleagues, we have discovered several reasons for why irrational decisions, as defined by economic models, can help leaders build trust. So far, we have relied on economic models to decide what to call rational. But if such rational decisions undermine trust, can we really call them rational? Why should economists have a monopoly on deciding what it means to be rational?

In newer work, my colleague Tessa Charlesworth (Northwestern University) and I have explored how non-experts interpret what it means to be rational. To do so, we use cutting-edge natural language processing tools, called word embeddings, to analyze how people use the word rationality in everyday life. We use these tools to analyze more than 840 billion words of Internet text.

Our first key finding was that people have a broader conception of rationality than is assumed by economic models. When people call someone else rational, they don’t just mean that they’re logical or smart; they also mean that they are trustworthy and reliable. This usage contradicts the classic portrait of the rational decision-maker as a Spock-like cold and calculating machine. Instead, it is consistent with models of rationality that emphasize the importance of cooperation for group success. 

But we didn’t stop there. Once we discovered these underlying dimensions of rationality, we could then use them to predict group stereotypes – and even demographic representation across different professions. Our results revealed that social groups (eg., men vs. women) are stereotyped along different dimensions of rationality and these stereotypes can predict representation of these groups across different professions. While this work does not yet have specific recommendations for managers, we believe managers and employees alike can take away a simple message: when trying to be rational, it is important to emphasize not just cold calculations, but also cooperation with others.