This is the fourth essay in Reckoning with Reality. The previous essay examined how intelligent people use competence to escape. This one asks the next question: once we leave the study, what kind of action actually brings us closer to the facts?
“Think less, do more” may be one of the hardest pieces of advice to dispute today.
It sounds brave and direct, and it often works. Many people are stuck not because they lack ability, but because they never quite put their work into someone else’s hands.
Yet I have grown wary of the phrase. Thinking can become an escape. So can action.
After that dinner conversation, the easiest thing for me to do was to turn the paths my friend mentioned into a plan immediately, using the feeling that I had “already started” to quiet my doubts. Looking back a few days later, I saw that the task list did not fill the real gap: a rule that would allow the result to change my direction. If every outcome could be explained as a reason to “keep going a little longer,” then moving faster would only bring me closer to my original belief, not necessarily closer to the facts.
Busyness Offers a Cheap Sense of Certainty
Action relieves anxiety quickly. Once we begin, our attention shifts from “Did I choose correctly?” to “What did I finish today?”
That shift has real value. Many uncertainties reveal themselves only after we start. The problem is that completing tasks feels so good that we can forget what we meant to test in the first place.
Publishing a hundred pieces of content, visiting fifty customers, and building twenty features are all countable actions. The larger the number, the more it resembles hard work—and the harder it becomes to stop.
But effort proves only that resources were spent. It does not prove the direction was right, or that we learned anything from the result.
Repeatedly raising the stakes at a casino is action too. The larger the bet, the stronger the emotion; the quality of the decision may be deteriorating all the while.
An Experiment That Cannot Fail Proves Nothing
The most useful part of scientific thinking has little to do with white coats or sophisticated statistics. Its essential feature is the willingness to let a result prove the original idea wrong.
Apply that requirement to ordinary decisions and many “experiments” are exposed as impostors.
If people buy the product, demand is real; if they do not, the market needs more education. If the content gets views, the direction is right; if it does not, the sample is too small. Whatever happens, the conclusion is to continue.
Actions like these cannot produce new judgment. Reality is permitted to supply material, but not to veto the plan.
Writing down “What result would make me stop?” before beginning sounds pessimistic. In fact, it protects the information value of the experiment. A stopping rule gives failure meaning, and prevents us from rewriting the standard after we have invested in the outcome.
When I evaluate an action now, I ask: What conflicting explanations could account for the result? Which of them can this action eliminate? If the opposite result appears, what am I prepared to change?
If I cannot answer all three, the action is probably only an expression of resolve.
The Size of the Bet Should Match the Strength of the Evidence
People like stories of success after a person goes all in because they are dramatic. Someone wagers all their savings, endures enormous pressure, and finally turns their life around.
Such stories naturally omit the failures. Those who lost have no company, stage, or interview from which to explain how the same kind of courage ruined them.
Survivorship bias is most dangerous when it turns an outcome into a character trait. The winner’s all-in bet is called foresight; the same move by someone who fails is called recklessness.
At the moment of decision, the evidence available to both may have been identical.
A sturdier rule is this: when evidence is weak, cap the downside first; as evidence strengthens, increase the commitment in stages. “Small” means more than a small sum of money. It also means easy to stop, limited in liability, and able to preserve choices after failure. When an action is cheap, reversible, and quick to yield feedback, we can think less and move faster. When a decision is expensive, difficult to reverse, and takes years to judge, we should first carve out a reversible stretch of road.
That is not timidity. Small bets allow a person to try more often while preserving the ability to invest more when the evidence turns favorable.
Some opportunities, of course, reveal themselves only after commitment. A partnership, a long research project, or a market entered through reputation cannot be understood fully from the sidelines. In such cases, there is no need to pretend that everything can be validated in advance. Commitment can be layered. Take on the portion that buys essential information first; postpone irreversible commitments of capital and identity.
Good Action Is Asymmetric
A worthwhile experiment usually has something like the structure of an option: I retain the right to walk away, my loss is capped, and a good result leaves the upside open. Remove any one of these features and “optionality” may become little more than a flattering name for risk-taking.
Testing a demand for two weeks with a fixed budget limits the cost of failure; if a strong signal appears, the test can continue. Serving one customer manually does not require building an entire product; if the same problem recurs, the experience, case study, and process all remain.
The return from such action is not limited to money. Even when it fails, it can eliminate a false hypothesis, expose a capability gap, or leave behind transferable relationships and knowledge.
Other actions have open-ended downside and limited upside. Borrowing heavily to prove one’s commitment to an unfamiliar industry can destroy the ability to try again if the judgment is wrong; even success may be consumed by high interest and fragile cash flow.
It is not enough to ask what an action could deliver in the best case. We must also ask whether the worst case is bearable, what survives failure, and whether there will still be a next choice.
Negative Knowledge Is Cheaper—and More Reliable
We want to know what is certain to work. Reality rarely gives us such answers. It is better at showing us what does not work.
One sale cannot prove durable demand. A set of rejections from different people that all point to the same obstacle has a better chance of overturning a hypothesis. But if ten rejections all come from the same channel, the same pitch, and the same type of person, they may prove only that the sampling method was poor. Negative evidence often depends on fewer assumptions than positive proof, but it is still constrained by sampling and experimental design.
This is the value of thinking in reverse. Instead of searching for a guaranteed path to success, begin by avoiding errors that can remove you from the game: unbearable debt, unit economics you cannot calculate, growth that requires rule-breaking, and a livelihood concentrated in one platform or customer with no way to migrate and no exit plan.
Early concentration is not necessarily foolish. One channel may accelerate learning; one customer may provide the wedge into an industry. The danger lies in rewriting temporary concentration as permanent safety.
Avoiding stupidity does not guarantee distinction, but it keeps us at the table. Many long-term outcomes do not require brilliance every time. They require us not to commit a small number of fatal errors.
The Product of Action Should Be the Next Judgment
I once treated delivery as the end of action. Build the product, publish the content, finish the process: task complete.
Now I think an action is unfinished until it has been reviewed.
Why did the result occur? Where was the original judgment wrong? Which signals were already present before we began but went unnoticed? What should we increase, reduce, or stop next time?
Without such questions, experience is merely something that happened. Once organized, it may become judgment.
The most valuable meaning of “just start” is not that speed should overpower thought. It is that thought should meet reality as soon as possible, then bring reality back into the next round of thought. Good action is a hearing that thought voluntarily arranges for itself.
Thought and action were never opposing camps. Poor thinking delays action; poor action refuses revision. At their best, the two form a loop that keeps getting shorter.
Action does not automatically lead to truth. Only action that permits failure, protects the next choice, and genuinely changes judgment deserves to be called an experiment.
Responses