How entrepreneurs navigate uncertainty

No founder gets to wait for perfect data. By the time a market signal becomes unambiguous, the opportunity may already have passed. Founders routinely make consequential decisions with partial information: when to hire, whether to raise funding and whether a shift in customer data is random noise or an early warning sign. Established companies may be able to draw on years of internal data and dedicated analytics teams. Early-stage founders rarely can.

Fortunately, some of the most influential economists and thinkers of the past century have spent decades studying how people make decisions when information is incomplete. Their findings do not offer a way to eliminate uncertainty. Instead, they show how people can make more deliberate and consistent decisions despite it.

Daniel Kahneman, a psychologist whose work with Amos Tversky reshaped modern economics, showed that people do not evaluate uncertain choices in the way classical economic theory once assumed. Rather than assessing outcomes only in terms of their objective value, people judge them relative to a reference point and tend to react more strongly to potential losses than to comparable gains.

“Gains and losses are short-term,” said Kahneman. “They’re immediate, emotional reactions. This makes an enormous difference to the quality of decisions.”

Kahneman also described two modes of thinking: a fast, intuitive mode that operates automatically, and a slower, more deliberate mode that requires greater effort. Founders often need fast judgment to keep a business moving, particularly when they are making dozens of decisions each day. However, consequential decisions, especially those that are difficult to reverse, may warrant more deliberate analysis.

The challenge is not to avoid intuitive thinking altogether. It is to recognize which decisions require extra scrutiny and to create enough time and structure to give them that attention.

Kahneman’s research also points to an uncomfortable pattern in how conviction can form.

“Subjectively it feels that you believe in that thing because you have the arguments for it,” he said. “That’s not the way it works. It works the other way around. You believe in the conclusion, and then you believe in the arguments that support the conclusion. That’s fundamental.”

For founders, this tendency can be particularly relevant when testing a new product, evaluating customer feedback or deciding whether a business strategy is working. Once someone becomes invested in an idea, it can become easier to notice evidence that supports it and discount information that does not.

One way founders can apply Kahneman’s insight is to build safeguards into the decision process before a strong conclusion forms. They might write down the evidence that would change their mind before beginning their research, assign someone the explicit task of presenting the opposing case or conduct a pre-mortem by imagining that a decision has failed and working backwards to identify possible reasons.

These practices cannot remove bias entirely. They can, however, make opposing evidence more difficult to ignore and encourage founders to test their assumptions rather than simply defend them.

Herbert Simon, a Nobel laureate economist and founding figure in decision science, studied how people and organizations make choices when their time, information and cognitive capacity are limited.

“I think the difficulty of decision-making centers very much around the degree of uncertainty and the gaps in our knowledge,” said Simon.

He also drew attention to the way choices may be constrained before the decision process has even begun. “Before you even start the choice process, somebody has presented you with this, and this, and this alternative,” he said.

Simon’s response to that problem became known as bounded rationality. No one, however capable, has unlimited time, information or cognitive capacity with which to compare every possible option and identify a mathematically optimal choice. Instead, people often establish a threshold for an acceptable option and stop searching when they find something that meets it. This is sometimes known as “satisficing.”

Simon viewed this as a realistic response to limited information, rather than simply a failure of rigor. His work also showed that organizations may generate more alternatives than they can meaningfully evaluate, while devoting insufficient structure to the choice itself.

A founder applying this principle could define what “good enough” means before beginning a search. For example, when comparing vendors, candidates or pricing approaches, the founder could identify the essential criteria in advance and establish a deadline for the decision. Once an option meets the agreed requirements, the search can stop.

This does not mean accepting the first available option or lowering important standards. It means deciding in advance which standards matter, rather than allowing the pursuit of an undefined perfect answer to delay action indefinitely.

Kahneman drew a sharp distinction between bias and what he called noise.

“There is variability, that variability is called noise. And the average error is the bias,” said Kahneman.

Bias describes a systematic tendency in a particular direction. Noise refers to unwanted variability between judgments that should ideally be consistent.

Founders can introduce noise without noticing it. A candidate may be assessed differently depending on the order of interviews. A pricing recommendation may shift according to who speaks first in the room. Two people may reach different conclusions from the same information because they have used different criteria.

Kahneman suggested that when several people are contributing to a decision, each person should record an assessment before the group begins its discussion. “If you have different people who are going to reach a decision together, the noise reduction technique is to have every one of them write down their answer before the discussion,” he said.

For hiring decisions, he also recommended evaluating candidates individually and assessing separate dimensions before forming an overall impression.

Founders can apply these insights through structured interviews, shared evaluation criteria and standardized scorecards. Evaluators can assess each dimension separately and submit their judgments before discussing the decision as a group.

These practices do not eliminate human judgment. They help prevent inconsistent judgments from being mistaken for meaningful differences.

Risk describes situations where the odds are knowable, even if the outcome is not. Uncertainty describes situations where even the odds are unclear, such as launching a genuinely new product category or responding to a geopolitical shock with few meaningful precedents.

Daniel Kahneman's research showed that people do not always evaluate such choices in a purely objective way. They tend to react more strongly to potential losses than to comparable gains, and the way a choice is framed can influence how its possible outcomes are perceived.

For founders, this means that uncertainty can sometimes feel more threatening than the available evidence justifies. A measurable risk can be analyzed using the information and probabilities available. Genuine uncertainty requires judgment because some of the information needed to calculate the odds may not yet exist.

Kahneman’s work suggests that a useful first step is to distinguish between what can realistically be estimated and what cannot. The goal is not to eliminate uncertainty, but to approach a decision with a clearer understanding of where analysis ends and judgment begins.

Vernon Smith, who founded the field of experimental economics, built his career around testing how people and markets behave under controlled conditions. Rather than relying only on theories about how markets should operate, his experiments examined what people actually did when faced with real choices and incentives.

“I think you always learn the most when you turn out to be wrong,” said Smith.

His research on market bubbles also highlighted the role of experience in changing behavior. Experienced traders did not necessarily avoid mispricing because they had reasoned their way to the correct answer in advance. Repeated exposure to market outcomes helped shape their decisions.

“They get there not by logic or rational thought, but by experience,” Smith said. “I think that’s the way most people in the economy act. What they believe about markets is driven by their experience.”

For founders, the broader lesson is to create opportunities to learn at a manageable cost. This makes reversibility a useful way to classify decisions.

A decision that can be changed relatively easily, such as testing a price, message or landing page, can often be treated as an experiment. It can be made more quickly because the result provides information and the decision can be adjusted.

A decision that is difficult or costly to reverse, such as choosing a co-founder, deserves more evidence, discussion and deliberate thought. The aim is not to give every decision the same amount of attention. It is to match the decision process to the potential consequences and the cost of changing course.

None of this research promises certainty. Founders will continue to make decisions without complete information, and no process can guarantee that every choice will produce the intended result.

What decades of research by Nobel laureates does offer is a set of practical habits.

Founders can distinguish between quick experiments and decisions that are difficult to reverse. They can define what “good enough” means before beginning a search, rather than allowing the pursuit of a perfect option to become its own form of paralysis. They can collect independent judgments before a group discussion and assess separate dimensions before forming an overall impression. They can also decide in advance what evidence would cause them to reconsider an existing belief.

When a decision feels obviously right, that feeling does not need to be dismissed. It can instead be treated as one piece of information that deserves to be examined alongside the available evidence.

Uncertainty is not a problem founders can eliminate. It is a condition they can learn to navigate. A more structured decision process cannot ensure the right answer, but it can make the reasoning behind a choice more deliberate, consistent and open to learning.