Why Hiring for Potential and Chasing Speed Fail for the Same Reason

Jul 27, 2026

10 min read

2 views

84%

The hidden trap in every decision made under uncertainty

What do a hiring loop and a research workflow have in common? More than most teams realize. In both cases, you are trying to make a high stakes judgment with incomplete information, and the temptation is to smuggle uncertainty into a shortcut. In hiring, that shortcut is potential. In research, it is the fantasy that progress is about reaching the destination first instead of understanding the map of possible routes.

Those shortcuts feel intelligent because they sound optimistic. Potential sounds generous, forward looking, and human. Speed sounds ambitious and competitive. But both can become excuses for vague reasoning. Once you start asking not just what you want, but how you know, the problem changes shape. The real challenge is no longer selecting the best person or the best method. It is designing a system that can make uncertain decisions without pretending certainty exists.

That is the deeper connection here: when the future is hard to see, the quality of your process matters more than the glamour of your prediction.

Why “potential” is a dangerous word

Hiring for potential has a seductive logic. If someone has not done the exact job before, surely we can identify their promise and place a smart bet. But the problem is that potential is often just a story we tell after noticing something we already liked. It is unusually easy to project competence onto people who resemble prior winners, communicate confidently, or match our unconscious expectations.

That makes potential one of the most common ways bias sneaks into supposedly merit based systems. The word sounds objective, but in practice it often means, “I believe this person could grow, and I trust my intuition about that growth.” Intuition is not worthless. The issue is that without a rubric, it becomes unexamined preference wearing a blazer.

This is why teams get trapped in inconsistent hiring. One interviewer sees potential in sharp opinions. Another sees it in quiet thoughtfulness. A third sees it in pedigree. A fourth sees it in speed. Each person feels rational, but the system is incoherent. You are no longer comparing candidates against a standard, you are comparing the candidate against each evaluator’s private imagination.

A useful test is simple:

If a hiring signal cannot be explained, observed, and scored in a way that different interviewers can independently reproduce, it is not a signal. It is a preference.

That does not mean you can never assess future growth. It means that if you do, you must define discoverable evidence of that growth. What does the candidate do that consistently indicates learning velocity, problem decomposition, or adaptability? What do those behaviors look like in a real interview, not in a vague impression? Unless you can answer those questions, you are not measuring potential. You are rewarding charisma, familiarity, or luck.

The real lesson is broader than hiring: whenever you say, “We are selecting for what this could become,” you have to ask whether you are protecting the decision from bias or merely disguising it.

Speed is not a race, it is a system property

The research quote offers a useful corrective to another common illusion: that progress is a straight line from idea to outcome, and whoever reaches the outcome first wins. In complex fields, especially medicine, the destination matters, but the route matters just as much. Knowing the landscape means understanding the options, the tradeoffs, the shortcuts that are actually dead ends, and the roads other people have already tested.

This is why “who gets there first” is often the wrong question. The better question is how quickly a team can move through the space of possibilities without becoming lost in it.

Speed, then, is not merely about rushing. It is about reducing unnecessary uncertainty. A fast team is not one that moves recklessly. It is one that knows where the bottlenecks are, which paths are promising, which assumptions are fragile, and which steps can be parallelized. In medicine, that might mean using AI to survey prior work, classify findings, map subfields, or surface overlooked connections. In a company, it might mean understanding which parts of a hiring funnel, product cycle, or decision process are slowing learning.

This perspective changes what we admire. The fastest teams are often not the ones with the boldest leaps, but the ones with the clearest decision architecture. They do not just ask, “How do we go faster?” They ask:

What do we need to know before acting?
What can be discovered cheaply?
What is irreversible?
Where are we confusing motion with progress?

That last question is crucial. A team can be busy, even impressive, while learning almost nothing. Real speed is the rate at which useful certainty accumulates.

The shared problem: decision making under fog

At first glance, hiring and research seem like different worlds. One is about choosing people, the other about choosing paths to knowledge. But both suffer from the same structural problem: you must act before the full truth is available.

That is where many organizations fail. They try to eliminate uncertainty by pretending their judgments are more objective than they are. In hiring, they lean on vague ideas of “fit” or “potential.” In research and innovation, they lean on the prestige of moving quickly or being first. In both cases, the hidden assumption is that the right instinct will compensate for the wrong process.

It usually does not.

The deeper principle is this: when uncertainty is high, process is the product. If your selection process is biased, your team composition will drift toward sameness. If your discovery process is sloppy, your speed will become noise. The way you decide is not separate from what you get. It is the mechanism that creates the outcome.

Think of it like navigating in fog. A bad navigator claims to know exactly where the destination is and walks confidently in the wrong direction. A good navigator does something less glamorous and much more effective: they check bearings, mark landmarks, compare routes, and update based on evidence. They are not obsessed with being clever. They are obsessed with staying oriented.

This is what makes “potential” and “speed” such revealing concepts. Both can be useful, but both become dangerous when they substitute for orientation. Potential without rubric becomes projection. Speed without landscape becomes haste.

The goal is not to make uncertainty disappear. The goal is to make uncertainty legible enough that your decisions improve inside it.

A practical framework: from intuition to evidence to acceleration

A strong organization does not choose between caution and ambition. It builds a sequence in which caution makes ambition possible.

Here is a useful framework for any domain where the stakes are high and the future is unclear:

1. Define the outcome you actually care about

This sounds obvious, but many teams confuse proxy goals with real goals. A hiring team may say it wants “potential,” when what it really wants is learning agility, independent judgment, or future leadership. A research team may say it wants speed, when what it really wants is faster discovery with lower error rates.

Be precise. Vague goals create vague evaluation.

2. Convert aspiration into observable behavior

If you value potential, identify what it looks like in action. For example:

Does the candidate improve their thinking when challenged?
Can they decompose a messy problem into manageable parts?
Do they learn from feedback within the interview itself?

If you value speed in research, specify which behaviors increase it:

Rapid literature scanning
Better synthesis of prior work
Fewer redundant experiments
Faster identification of dead ends

The point is not to reduce human judgment to a spreadsheet. The point is to make judgment accountable to evidence.

3. Build a rubric that others can use independently

A rubric is not bureaucracy. It is a defense against private narratives. If multiple people can apply the same criteria and broadly agree, your standard is probably real. If they cannot, your standard is probably elastic.

This matters because organizations often mistake consensus for objectivity. In truth, consensus can simply mean that everyone shares the same unspoken bias. A good rubric forces disagreement into the open where it can be examined.

4. Optimize for learning rate, not just outcome

This is where the two source ideas finally converge most powerfully. A hiring loop should not merely ask, “Did we pick someone who succeeded?” It should ask, “Did our process help us identify useful signals, or did it rely on impressions that could not be repeated?”

Likewise, a research system should not only ask, “Did we arrive first?” It should ask, “How quickly did we learn which paths were worth pursuing?”

In both cases, learning rate is the hidden metric. Organizations that learn faster make better decisions over time, because they continuously improve the quality of the questions they ask.

The best systems do not worship instinct, they train it

There is an important subtlety here. This is not a call to eliminate human judgment. In messy domains, judgment is unavoidable. The mistake is to confuse judgment with unstructured intuition.

A well designed hiring process does something more sophisticated than “remove bias.” It channels intuition into evidence. A good interviewer may still feel strongly about a candidate, but that feeling must be anchored in specific, visible signals. Similarly, a good researcher may feel drawn to a promising direction, but the feeling must be tested against the field, the alternatives, and the actual bottlenecks.

This is how expertise matures. Experts are not people who always know the answer immediately. They are people whose intuition has been trained by repeated contact with reality, and whose judgments can be articulated when challenged. That is why the most reliable experts are often the least mystical. They know which parts of their judgment are solid and which parts are provisional.

A mature organization borrows that humility. It says, in effect:

We can value promise, but only if we can name the behaviors that reveal it.
We can value speed, but only if we can map the landscape well enough to avoid false shortcuts.
We can trust instinct, but only after we have made it testable.

That is a much stronger position than either blind rigor or blind optimism. It respects human judgment without allowing it to go unchecked.

Key Takeaways

Do not hire for “potential” unless you can define it behaviorally. If two interviewers cannot independently score the same evidence, the signal is too fuzzy to trust.
Treat speed as a rate of learning, not a race to brag about. The fastest teams are the ones that reduce uncertainty the quickest.
Use rubrics to protect judgment from projection. A rubric turns private impressions into shared criteria.
Ask what landscape knowledge would change your path. In any complex project, map the alternatives before optimizing for momentum.
Measure process quality, not just outcomes. Good outcomes can come from bad systems, but bad systems eventually produce bad outcomes.

The real competitive advantage is not certainty, it is clarity about uncertainty

The temptation in every high stakes decision is to seek a shortcut that makes uncertainty feel smaller than it is. In hiring, that shortcut is the myth of seeing hidden potential. In research and innovation, it is the myth that speed alone wins. But both collapse under the same pressure: reality eventually demands that you explain how you knew.

That is why the strongest organizations are not the ones that pretend to see the future. They are the ones that build methods for moving responsibly through fog. They know that fairness, speed, and quality are not separate virtues. They are different expressions of the same discipline: making better decisions before all the information arrives.

So perhaps the question is not whether you should hire for potential or chase speed. The more revealing question is this:

What would your process look like if you cared more about making uncertainty visible than about sounding confident in it?

Once you can answer that, you are no longer just selecting people or accelerating research. You are designing an organization that learns faster than its own ignorance.