Dan Herbatschek believes most people have never examined how they make decisions, and that kind of oversight is costly. He has watched capable teams arrive at poor outcomes from a lack of a coherent framework. A smart decision is a recognizable architecture, built from how a problem is framed, what information is gathered, how uncertainty is handled, and what time horizon governs the whole.
Framing: The Step That Shapes Everything Else
Every decision begins before the options are visible in the act of framing. Framing determines which solutions surface and which never appear, encoding assumptions that govern every subsequent step.
The danger is that framing errors are invisible to the person who commits them. Once a problem is described a certain way, the mind works inside that description rather than against it. Organizations are especially vulnerable, preserving inherited frames long past their usefulness.
“Most decision failures I’ve observed weren’t failures of analysis,” Herbatschek says. “They were failures of framing. The team solved the problem they were given, but nobody stopped to ask whether the problem they were given was the real one.”
The corrective is simple. Before investing in analysis, interrogate the question itself.
Information: How Much Is Enough, and What Kind Matters
Once a problem is properly framed, the question becomes informational. The instinct in data-rich environments is to gather more, to delay until the picture is complete. That instinct is often wrong, as the goal is not comprehensive information but sufficient information that bears directly on the specific choice at hand.
Dan Herbatschek draws on his machine learning background to distinguish signal from noise. A signal is information that changes what you would do. Noise feels relevant, but wouldn’t alter the decision regardless.
Most professionals spend the majority of their analytical energy on noise because it is more available and less threatening than the signal that actually demands a response. The discipline is to identify the choice first, then ask only what information would change it.
Uncertainty: Working With What You Cannot Know
No decision of any consequence is made in conditions of perfect certainty. Uncertainty is simply the environment in which good decision-making happens. The question cannot be how to eliminate it, because that is impossible. Instead, the question must be framed as how to engage with it honestly, and how to make choices that remain sound across a reasonable range of possible futures instead of only in the single future the decision-maker happens to prefer.
Herbatschek sees uncertainty the way a well-designed system would approach variance, by building robustness into the structure. A decision that is highly sensitive to small changes in assumptions is a fragile decision, regardless of how well-reasoned it appears under favorable conditions.
Identifying those sensitivities by stress-testing the conclusion against the most likely ways the underlying assumptions could be wrong is structural integrity.
“Uncertainty isn’t the enemy of good decisions,” Herbatschek notes. “Pretending it doesn’t exist is. The teams that make consistently strong calls are the ones that name their unknowns explicitly, rather than quietly assuming them away.”
Naming uncertainty out loud changes the character of deliberation. It shifts the group from a posture of proving the decision right toward the harder, more honest work of understanding the conditions under which it might be wrong and deciding if those conditions are tolerable.
Values and Time Horizon: The Two Variables Most Often Left Out
Analysis and information can only carry a decision so far. At some point, every meaningful choice involves values and judgments about what matters when genuine trade-offs cannot be resolved by data alone. Most professional decision-making cultures are uncomfortable with this.
Values feel subjective, and subjectivity feels like weakness in environments that prize analytical rigor. The result is that value judgments get made implicitly, buried inside the framing or embedded in the weights assigned to competing criteria, without ever being examined or acknowledged as such.
Making values explicit makes a decision more honest. When a team names what it actually cares about, the decision-making process becomes more coherent and more auditable. Disagreements surface in the deliberation where they should as opposed to in the aftermath. And the decision itself becomes something the team can genuinely defend, because it rests on principles that were chosen consciously.
A decision that is optimal over six months may be destructive over five years, and vice versa. Smart decision-making requires naming the relevant time horizon explicitly and testing if the preferred option holds up across it. Herbatschek has seen this conflation produce consistent damage in organizational settings.
“A lot of what looks like strategic disagreement is actually a disagreement about time. Two people can look at the same evidence, apply the same logic, and reach opposite conclusions, simply because one is optimizing for next quarter and the other is optimizing for the next decade,” says Herbatschek.
The Integrity of the Process
Outcomes are partially determined by factors outside any decision-maker’s control, and a sound process can still produce a bad outcome when circumstances intervene. The inverse is equally true, as a poor process can produce a good outcome through luck.
Conflating the quality of the decision with the quality of the result is one of the most persistent errors in how organizations learn, or fail to learn, from experience. What makes a decision genuinely smart is the integrity of the process that produced it.
The framing was honest, as the information gathered was sufficient and relevant. Uncertainty was named instead of suppressed. The values at stake were made explicit, and the time horizon was defined and defended.
A decision built on those foundations may still fail when the world behaves unexpectedly, but it gives the people who made it something far more valuable than a good outcome. It provides a process they can examine, refine, and trust the next time a hard choice arrives. And in a world where hard choices arrive constantly, that process is the only durable advantage there is.
