Preventing overconfidence from undermining your practice

We all know someone convinced they know more than they do. The hard truth is each of us is guilty of overestimating our cognitive prowess. We can't help it. Overconfidence is hardwired into us. It's a failure in self-awareness tied to an inherent bias that protects our ego by convincing us we are smarter than we actually are.

Social psychologist David Dunning, PhD, identified this phenomenon, known as the Dunning-Kruger effect, in a paper co-authored with Justin Kruger and published in a 1999 issue of the Journal of Personality and Social Psychology.

It is based on research in which participants were tested in areas such as humor, grammar, and logic, and then they predict their performance relative to other participants. The findings showed the lowest performers scored their performance far higher than they achieved while the highest performers tended to slightly underestimate their performance.

Ultimately, Drs. Dunning and Kruger discovered that people often lack the skill to accurately assess their competence and that confidence is more often a result of ignorance than knowledge.

"Those who lack expertise lack the expertise to know that they lack expertise," Dr. Dunning said.

Dr. Dunning spoke about how this phenomenon hamstrings businesses during the 2025 AVMA Veterinary Business and Economic Forum, held October 8-9 in Denver. Additionally, he suggested tips that practice leaders can use to account for the Dunning-Kruger effect and mitigate its impact on operations.

The bane of overconfidence

Since the paper's publication, the Dunning-Kruger effect has been used to identify gaps in perceived versus actual competence in fields as diverse as computer programming, investing, or medicine.

So, when choosing a diagnosis, a pricing strategy, or a staffing plan, we select the option that seems most reasonable to us, and therefore it feels right, Dr. Dunning explained. However, our internal feeling of rightness is a poor proxy for truth. For example, in a laboratory task diagnosing "zombie diseases," novices gained confidence faster than skill, producing a "beginner's bubble" in which errors spike after a bit of experience, not on Day 1. Dr. Dunning noted similar real-world patterns: private pilots enter a higher-risk window after tens to hundreds of hours, and surgeons learning a new procedure make more errors after their first 10 to 15 cases.

"Not knowing what you don't know might be in the brain, but ego might be a factor somewhere in this, too," he said.

Overconfidence is the most consequential of human biases and "the one to get rid of" if it were possible, according to Dr. Dunham. Unfortunately, human beings are hardwired to be sure of themselves, even when wrong. "We underestimate how much we don't know, assume our judgments are more accurate than they are, and fail to recognize how unpredictable the world really is. For businesses, that combination of misplaced certainty and limited perspective can be devastating," he said.

It manifests as misestimated demand for a new service, narrow scheduling buffers that implode when procedures run long, premature autonomy granted to new hires, and leaders assuming their instructions were clear when they weren't.

Entrepreneurs are especially prone to this kind of thinking, Dr. Dunning said. Every founder believes they're the exception to the statistics, convinced that hard work or passion will spare them the fate of the majority. Yet the Bureau of Labor Statistics shows that 20% of new businesses fail in their first year, about half within five years, and roughly two-thirds within 10 years.

It's not that most entrepreneurs are reckless, Dr. Dunning said, they simply overestimate how much they understand their market or how unique their idea is.

A similar effect drives what economists call the "winner's curse," which is when the highest bidder for a property or contract is also the one who most overestimated its value. "The highest bidder is often the one most wrong," he said. "Confidence pushes them past prudence."

Even technical fields full of trained experts fall into the same trap. Dr. Dunning showed data from chemical engineers and software developers, most of whom rated their performance in the top 5% of their peers, which is a mathematical impossibility. Experience tends to reduce the number of mistakes, he said, but it doesn't necessarily make people humbler. Confidence stays high, even as accuracy only improves a little. The same pattern appears in financial markets, product design, and project management.

"It's not that people don't learn. They learn, but their confidence learns faster than their competence," he noted.

The fix

Dr. Dunning contrasted these patterns with professions that have developed safeguards against overconfidence. Medicine, he noted, relies on differential diagnosis, requiring clinicians to consider multiple possible explanations before settling on one. The legal system achieves something similar through adversarial arguments, where opposing sides test each other's assumptions before a neutral judge or jury. Both systems institutionalize doubt.

Business planning, in contrast, often assumes that the loudest idea is the best one.

To manage overconfidence, Dr. Dunning recommended adopting habits from those self-correcting fields. Leaders should make a point of considering alternatives rather than rushing toward confirmation. Scientists, he said, don't design experiments to prove their theories right; they design them to see if they can survive being proved wrong. Businesses can do the same by actively testing assumptions and seeking disconfirming evidence.

Additionally, Dr. Dunning urged managers to "take the outside view" by anchoring plans in historical data, such as base rates for similar projects or ventures rather than treating every new initiative as unprecedented.

Another defense is to build buffers directly into operations.

"Microsoft says, 'Three months? Terrific—four,'" Dr. Dunning said, explaining the need to plan for the human tendency to underestimate. Architects often double their material estimates for the same reason, and veterinary practices can do it by padding implementation timelines, preserving cash reserves, or overhiring slightly during busy seasons.

When failures occur, Dr. Dunning suggests leaders analyze them using Toyota's "Five Whys" method, asking "why" five times to find the root cause rather than blaming the nearest person.

He suggested that before launching new services or projects, businesses conduct a "pre-mortem," that is, imagining that the initiative has failed and working backward to determine how and why. The exercise forces teams to think of possibilities they would otherwise overlook.

For managers, Dr. Dunning advised when giving feedback to focus on what can be changed. Do not make it personal, do it frequently and especially when it's low stakes, and be sure to point out the positive. Finally, he said, come up with a plan going forward and make sure the employee can articulate it.

In the end, overconfidence is not a moral failing but a cognitive fact, Dr. Dunning explained. "Our genius is essential, it gets us through novel situations, but sometimes it misfires," he said. The goal for business leaders is not to avoid all mistakes but to build organizations resilient enough to absorb them.

"Overconfidence is inevitable," Dr. Dunning concluded, "but resilience just concedes that fact and plans around it."