The Jail-Cell Test: How to Measure High Agency

High agency is the single trait that most reliably separates founders who shape outcomes from those who absorb them. George Mack’s jail-cell test — if you woke up in a foreign jail cell and could call only one person, who would it be? — operationalizes that trait in under ten seconds. The harder, more productive version of the question is the one most people skip: would anyone call you?

Why the jail-cell test works as a diagnostic

You wake up in a foreign jail cell. You are allowed to call only one person to get you out. Who do you call? The scenario is artificial, but the cognitive work it forces is real. You are not selecting the most intelligent person you know, or the wealthiest, or the most credentialed. You are selecting the person most likely to actually solve the problem — someone who will not freeze, will not defer, and will not wait for a more convenient moment. The one thing everyone in that high-agency room has in common: they are happening to life. They do not view the future as a static entity. They view it as something to be shaped by human action.

The test was popularized by writer and entrepreneur George Mack, who created the viral highagency.com — explaining high agency in 30 minutes. Its elegance is that it bypasses résumé signals entirely. If you created a room with everyone you would call when stuck in a foreign jail cell, what would that high-agency room have in common? It is not age, gender, race, education, job titles, or politics. It is not optimism or pessimism either. The common thread is a disposition toward action under constraint — which is, of course, the operating condition of every founder, operator, and investor worth the title.

First coined in 2016 by venture capitalist Eric Weinstein, who linked the “high-agency” mindset to a “MacGyver” approach to corporate problem-solving, the term was used to describe especially driven entrepreneurs. Startup adviser Shreyas Doshi defines it this way: “High agency is about finding a way to get what you want, without waiting for conditions to be perfect or otherwise blaming the circumstance.” The concept has since migrated from niche VC vocabulary into mainstream business discourse: analytics site Brandwatch found there was a 500 percent jump in mentions of the phrase across X, Reddit, and other social media sites. The jump in mentions is a symptom of a real need — founders and operators are searching for a vocabulary that captures something they observe in the best people around them but struggle to name precisely.

This article gives that vocabulary structure. It breaks the jail-cell test into three measurable components, maps each to a body of evidence, and closes with drills you can run this week.

Component one: clear thinking under stress

The person you call in a crisis is not necessarily the smartest person you know in a seminar room. They are the person who can think clearly when the environment is hostile, information is incomplete, and the clock is running. This is a distinct cognitive skill — one that degrades sharply under panic and improves with deliberate practice.

Mack identifies a telling behavioral signal: before rushing to answer your question, high-agency people question whether it is the right question to answer. They know the right answer to the wrong question is worse than no answer to the right question. This is not pedantry. It is the discipline of refusing to optimize a bad frame — the single most expensive mistake a founder can make in a crisis.

The psychological substrate here is well-documented. The concept of “high agency” is an amalgamation of, or an umbrella term for, a range of traits that psychologists have studied for decades. Related concepts include the prized “growth mindset,” “proactivity,” and the somewhat controversial “grit.” More precisely, the research literature points to internal locus of control as the load-bearing variable. Internal locus of control indirectly affects venturing outcomes via entrepreneurial competency, whereas external locus of control has no such consequences. An internal locus of control is associated with a greater sense of responsibility for one’s decisions. Entrepreneurs with this mindset are more likely to carefully analyze situations and take calculated risks. They are less likely to blame external factors for their failures, instead learning from their mistakes to make better decisions in the future.

Clear thinking under stress is also visible in how high-agency people handle information. A low-agency trap is to be hypnotized by groupthink. High-agency people refuse to passively download the current consensus without first verifying it for themselves. In practice, this means they arrive at decisions with a first-principles map of the situation rather than a borrowed one — which is precisely what you need when the map everyone else is using has stopped working.

Drill: the five-minute pre-mortem

Before any significant decision, write down the single most likely reason it will fail. Not a list — one reason. Force yourself to name the highest-probability failure mode, then ask whether your current plan addresses it. If it does not, you are optimizing the wrong variable. Repeat weekly until the habit is automatic.

Component two: bias to action

The person you call from a jail cell is not someone who will research the problem extensively and schedule a follow-up. They will move. Paul Graham distilled being a good startup founder to two words: relentlessly resourceful. Till then the best he had managed was to get the opposite quality down to one: hapless. Haplessness, in Graham’s framing, is not bad luck — it is passivity dressed as circumstance. Hapless implies passivity. To be hapless is to be battered by circumstances — to let the world have its way with you, instead of having your way with the world.

Bias to action is not recklessness. It is the capacity to make a reasonable decision with available information and then move, rather than waiting for certainty that will never arrive. Naval Ravikant says that when building teams, he looks for people who “just solve problems without even being asked.” These individuals identify what needs to be done and handle it — they are not constantly asking for permission or basic guidance.

Keith Rabois, the investor and operator behind PayPal, LinkedIn, and Square, captures the organizational dimension of this with his barrels-vs-ammunition framework. If you think about people, there are two categories of high-quality people: there is the ammunition, and then there are the barrels. You can add all the ammunition you want, but if you have only five barrels in your company, you can literally do only five things simultaneously. If you add one more barrel, you can now do six things simultaneously. Finding those barrels — someone who can take an idea from conception to live and it is almost perfect — are incredibly difficult to find. This kind of person can pull people with them. They can charge up the hill. They can motivate their team, and they can edit themselves autonomously.

The jail-cell test, reframed: the person you call is a barrel. The question is whether you are one.

Mack identifies a behavioral tell that distinguishes barrels from ammunition in real time: when they ask for your help, they respect your time by asking specific questions and avoiding vague ones. To show appreciation for the time given, they follow up with freakish speed with a thank you and evidence of the advice actioned. Speed of implementation is not a personality quirk — it is a signal that someone has internalized the cost of delay.

Drill: the 48-hour rule

For every piece of advice, insight, or decision you receive this week, implement at least one concrete action within 48 hours. Not a plan to act — an action. Track your hit rate. Most intermediate founders discover they are implementing fewer than half of their own decisions within a week. The gap between decision and action is where agency leaks.

Component three: productive disagreeability

This is the component most people omit from their mental model of high agency, and it is arguably the most important one for founders specifically. The person you call from a jail cell is not someone who will tell you what you want to hear. They will tell you what is true, even if it is uncomfortable, and they will push back on the authorities standing between you and the exit.

Mack describes this as a defining trait of high-agency people: the social incentives are to be nice to people’s faces and gossip behind their backs. To do the opposite requires agency because they are swimming against the social tide. This is not contrarianism for its own sake. It is the willingness to say the accurate thing when the social cost of saying it is real.

The research on this is striking. A paper published in Strategy Science examined venture competition data and found that the more venture competition judges disagree on the quality of a startup, the more likely the startup is to succeed, particularly when its proposition is unique. Entrepreneurs pursue opportunities based on their subjective beliefs, and common opinion alone cannot be a source of competitive advantage. Value is disproportionately created and captured by founders with unconventional ideas that spark disagreement. Productive disagreeability, in other words, is not a social liability — it is a leading indicator of genuine differentiation.

The distinction that matters is between productive and destructive disagreement. Productive conflict centers on the intellectual merit of ideas, while destructive conflict targets the individual’s character. High-agency operators are fluent in the former and allergic to the latter. They challenge the idea, not the person — and they do so in real time, not in the corridor afterward.

While high agency may seem like an innate personality trait, emerging research suggests the people around us may be a powerful source of personal agency. People who are better able to influence their own outcomes are often those who can turn to, or recruit, others to help them achieve those outcomes. Paradoxically, this means that “high agency” might not only be a quality of you personally, but a quality of the people around you. This is why the jail-cell test has a second, underappreciated dimension: the quality of your network is a proxy for your own agency. If you cannot name anyone you would call, the diagnostic is pointing at a gap in your relationships, not just your character.

Drill: the honest-room audit

Identify three people in your professional life who have told you something you did not want to hear in the last 90 days. If you cannot name three, you are operating in an echo chamber. Actively seek out one person this week whose default is to push back on your assumptions — not to be difficult, but because they care more about the outcome than about your approval.

Running the full self-diagnostic

The jail-cell test is most useful when applied symmetrically: first outward, then inward. The outward application is easy — most people can name their one call within seconds. The inward application is the diagnostic work.

Ask yourself the following questions honestly, and score each on a scale of one to five:

• Clear thinking under stress: In the last significant crisis you faced, did you reframe the problem before acting, or did you react to the first frame that presented itself?
• Bias to action: What is the median time between a decision you make and the first concrete action you take on it? Is it hours, days, or weeks?
• Productive disagreeability: In the last month, how many times did you say something accurate but uncomfortable to someone who had the power to make your life harder?
• Network quality: If you listed the five people you speak to most about your work, how many of them would you call from a jail cell?
• Reciprocity: Are there people who would call you? What is the evidence?

A score below three on any dimension is a development priority, not a character verdict. Research shows this psychological trait responds to training and experience. The components of high agency are not fixed — they are habits, and habits are built through deliberate repetition under real conditions.

Why high agency compounds

The reason high agency matters disproportionately for founders is not that it helps in individual crises — it is that it compounds across time. A founder who acts on 80 percent of their decisions within 48 hours will run roughly ten times more experiments per year than one who acts on 20 percent. More experiments means faster learning, faster iteration, and a compounding advantage in market understanding that no amount of strategic planning can replicate.

Research has identified that traits most strongly correlated with entrepreneurial success include a proactive personality and an internal locus of control, alongside a need for achievement and generalised self-efficacy. These are not independent variables — they are mutually reinforcing. A founder who acts more frequently gets more feedback, which sharpens their thinking, which improves the quality of their next action. The loop is self-reinforcing in both directions: high agency accelerates learning, and learning reinforces agency.

The inverse is equally true. People with an internal orientation take corrective action after setbacks. Those with an external orientation blame circumstances and stagnate. Stagnation is not a stable state in a startup — it is a slow decline dressed as a plateau. The founders who survive long enough to build something durable are almost always the ones who treat every setback as a data point rather than a verdict.

Mack’s observation about high-agency people and self-taught learning is relevant here: whether it is learning to play their favourite song on the saxophone or deconstructing how 3D printers work, they start from zero and use agency to climb up the knowledge ladder. The domain is irrelevant. The pattern — identify a gap, act to close it, iterate — is the same one that drives company-building. High-agency people practice it constantly, in every context, which is why they are so effective when the stakes are highest.

What this means

The forward view

The jail-cell test will not appear on any competency framework or performance review. It is not a hiring rubric or a board metric. It is something more useful: a mirror. It shows you, in a single question, whether you have built the habits, the relationships, and the disposition that make you genuinely useful when conditions are worst.

The founders who build durable companies are not the ones who avoid hard situations. They are the ones who have practiced being the person someone calls when a hard situation arrives — and who have built teams of people who are equally callable. That is not a personality type. It is a system. And systems, unlike personalities, can be designed.

The question is not whether you are high-agency. The question is whether you are building the habits that will make the answer yes when it matters most.

If you are working on building systems that compound — in hiring, capital, or credibility — the Six Gaps framework and the trust development model on FounderWise are natural next reads. The Business Growth Accelerator (a FounderWise brand) applies these frameworks in live operating contexts.