This governs how you make the same kind of decision twice. Dalio's claim is that a principle is narrower than a value: a value is what you care about, a principle is the reusable rule that connects it to action in a recurring situation, and if you have not written yours down you are either reacting to whatever shows up or living on someone else's rules untested. The whole document is one loop applied first to yourself and then to a group: set a goal, find the problem in the way, diagnose it to a root cause, design around it, and do the work.

Do

Write the rule down the moment a problem repeats. Dalio built roughly two hundred principles one at a time, by recording the type of problem he or someone he cared about had run into and the principle that had been violated to cause it. The number was never planned, it accumulated, and the organizing structure was imposed afterward to make it readable. [Source: "What Principles Are, and Why They Matter"]

Work the five steps as five separate exercises. Goals, problems, diagnosis, design, tasks. Dalio's rule is that blurring them, most commonly by thinking about solutions while you are still supposed to be diagnosing, degrades the output of every step at once. The related discipline is to hold goals stable and let designs and tasks flex underneath them, since most people who struggle have this reversed. [Source: "Your Two Yous and Your Machine"]

Set the goal before you assess whether you can reach it. Dalio says plainly that every time he sets goals he has no idea yet how he will achieve them, and treats ruling out a goal on a superficial read of its attainability as a mistake, because real feasibility only appears after the design step. He adds an explicit permission that school never gave you: you are allowed to recruit other people's expertise rather than needing every answer inside your own head. [Source: "Step 1: Setting Goals"]

State the problem precisely, with names, and never confuse it with its cause. Not "people don't like me" but who, and under what circumstances. Not the anonymous "we" or "they," which masks who is actually responsible, but a specific person. And "I can't get enough sleep" is a cause, not a problem: the problem is the poor performance it produces. [Source: "Step 2: Identifying and Not Tolerating Problems"]

Keep asking why until the answer stops being a verb and becomes an adjective. Dalio's own example: "I missed the train because I didn't check the schedule" names an action, which is a proximate cause. "I didn't check the schedule because I am forgetful" names a trait, which is the root cause and the only thing you can actually remove. Most people stop one question too early. [Source: "Step 3: Diagnosing the Problems"]

Design before you do. Dalio's estimate is that a good plan takes literally hours spread over days or weeks against the far longer time spent executing it, and that the single biggest mistake at this stage is skipping it entirely out of eagerness to start. The method is two-scaled and iterative: sketch the broad steps, fill them with specific dated tasks, and let the specifics force revisions back up to the sketch. [Source: "Step 4: Designing the Plan"]

Say it to the person, or don't say it at all. Bridgewater's rule is that you never say anything about someone you would not say to them directly, managers do not discuss a person's performance without that person in the room, and meetings are recorded and shared specifically so nobody can tell two audiences two different things. Dalio distinguishes this sharply from criticism, which is welcome, provided it goes to the face. [Source: "Trust in Truth"]

Weight opinions by believability, and ask for the reasoning rather than the conclusion. Believability is a rough measure of the probability someone is right, based on track record and the quality of their reasoning when probed, and Dalio's image is a novice golfer questioning Tiger Woods: worth doing, but not from a position of equal confidence. Before an important decision he triangulates with at least three believable people and asks each how they got there, so he can check the logic instead of borrowing the answer. [Source: "Constantly Get in Synch"]

Fight the small battles too. Dalio's stated policy is explicitly not to pick your battles but to fight them all, on three grounds: small badness is often a symptom of something larger, resolving a small gap in understanding now prevents a bigger one later, and each battle is itself a piece of training in the standard you actually want. Probing runs both directions, and past your direct reports to the people who work for them. [Source: "Probe Deep and Hard to Learn What to Expect From Your Machine"]

Don't

Don't tolerate a problem you have already seen. Dalio splits failure here into two kinds with two different fixes: not perceiving the problem, which needs sharper perception or outside eyes, and perceiving it and living with it anyway, which he treats as the more dangerous of the two. His four-stage scale runs unidentified, identified with no plan, identified with a good plan, solved, and he flags stage two as the worst for morale, because everyone can see the problem is real and nobody can see it moving. [Source: "Know How to Perceive Problems Effectively"]

Don't blame anything outside yourself for the outcome. Dalio's framing is that blaming a bad result on something external is essentially wishing reality were different than it is, and that it diverts attention from building the capability that would actually produce a better result next time. He pairs it with the related error of protecting how you look instead of getting better, which leaves the weakness permanently in place. [Source: "Your Most Important Choices"]

Don't lose the line from a task back to the goal it serves. The only purpose of a task is to achieve a goal, so Dalio's check is to keep asking why of everything on the list until the connection is visible again. People who struggle tend to be rigidly attached to specific tasks while quietly loosening their commitment to the goal those tasks were for. [Source: "The Relationships Between the Steps, and Why Weaknesses Don't Matter"]

Don't hire for the skill and hope the values follow. Dalio's ranking is values first, abilities second, skills a distant third, since skills are the easiest to acquire and the fastest to go obsolete. His standard is not "good enough" but a candidate who clearly stands out in their peer group, and he is candid that even after all this there is a high probability the person will not be the one you needed, which is why the assessment continues after the hire rather than stopping at it. [Source: "Hire Right, Because the Penalties of Hiring Wrong Are Huge"]

Don't soften an evaluation to be kind. Dalio's test for anyone already on the team is the same one used on a candidate: would I hire this person knowing what I now know. He calls it odd and silly that interviewers criticize outsiders freely and hesitate to apply the same scrutiny to their own people, and rejects the habit of balancing every criticism with a compliment, since the ratio should reflect reality rather than fairness. He does insist both go into explicit perspective, after a case where he praised a researcher, criticized him days later, and left the man convinced he was about to be fired. [Source: "Evaluate People Accurately, Not \"Kindly\""]

Don't order people around, and don't quietly do their work for them. His instruction is not to try to be followed but to be understood and to understand, because directives breed quiet resentment and covert defiance while genuine agreement about what is true makes people want the same outcome you do. The matching error is a manager who silently absorbs a failing subordinate's work, which destroys the only evidence of what that person can actually do: Dalio's phrase for the trade-off is crashing the car versus learning from the crash. [Source: "Manage as Someone Designing and Operating a Machine"]

Don't keep someone in a job that is not a fit. Dalio's position is that keeping a badly performing person is worse than firing them, because the person is blocked from real progress while the organization absorbs the underperformance and the erosion of its own standard. Redesigning the job around someone's weaknesses is the wrong fix, sorting them into a role that suits them or out of the company is the right one, and the bar itself does not move to accommodate anyone. [Source: "Sort People Into Other Jobs, or Remove Them"]

Don't put enough on any one decision that being wrong is ruinous. Every decision is an expected-value calculation of probability against payoff, with one hard constraint: the probability of an unacceptable outcome must be nil. Dalio's portfolio-style version of the rule applies well outside finance, make fifteen or more good uncorrelated bets, and he pairs it with a prioritization order that most people invert: get every must-do above an acceptable bar first, only then make them excellent, and only then start on the like-to-dos. [Source: "Make Decisions Effectively"]

Hold

Nothing improves while you are protecting how you look. Dalio names the defensive reaction to being shown wrong as the biggest single problem of mankind, not because people lack the capacity to reflect but because most will not aim it at their own weaknesses, and reaches for Aristotle's definition of tragedy: a bad outcome produced by a flaw the person cannot get around. Holding this makes an unflattering diagnosis usable information. Refusing it turns every accurate criticism into an attack that has to be repelled, which keeps the weakness exactly where it is. [Source: "My Most Fundamental Principles"]

You cannot stand still and stay happy. Dalio's account of happiness is comparative rather than absolute: it tracks outcomes against expectations, which is why a billionaire down $200 million is miserable and someone worth $10,000 who gains $2,000 is delighted. That leaves only two honest routes, raise expectations and keep exceeding them, or lower expectations to match your conditions. Holding this tells you which of the two you are actually attempting. Refusing it leaves you expecting a fixed achievement to produce a lasting feeling it has never once produced. [Source: "The Personal Evolutionary Process"]

You are one of your own resources, not the point of the exercise. Dalio splits a person into the designer who builds the machine and the worker who fills a slot inside it, and says the designer should be exactly as objective about that worker as about anyone else, up to and including taking them off a position they play badly. Holding this makes benching yourself an ordinary staffing decision. Refusing it makes every admission of a personal limit a defeat, which is why people keep occupying roles they are demonstrably wrong for. [Source: "Your Two Yous and Your Machine"]

The one line

Write down the rule the second time a problem happens, chase every failure back to a trait rather than an action, and be exactly as objective about yourself as you are about everyone you employ.