Chapter 3: The Age of Adaptive Insight: How Much You Know No Longer Matters

One or two decades #

Anyone who has ever done recruitment has encountered this moment: two resumes are placed on the table, both with "ten years of industry experience" written on them, and the asking prices are similar. After the interview, you know in your heart that these two people are not the same at all: when talking about any project, one can tell "what was the basis for judgment at the time, which one was later verified to be wrong, and what I would change today"; the other talks about ten years into a running account, and when asked about "why", the answer is "this is how it is done in the industry."

The same decade produced two such different people. Popular explanations are talent, hard work, luck, and platform. These explanations all hit a bit of a rough patch, but none of them hit the nail on the head. The mechanism is: There are two types of ten-year experience, one is ten years of standard accumulation, and the other is repeating the first year ten times. Every year of the former is producing verified judgment basis, while every year of the latter is just another step on the timeline. Working years record time; only verified standards can record growth.

In the old days, the difference between these two decades was blurred because knowledge and execution were valuable in themselves. Even if you have not set any standards, if you "know the industry practices" and you are "familiar", it is enough to support a decent premium. Chapter 1 has already explained the trends of these two premiums: they are both approaching zero. So a gap that could have been decently hidden in the past is now exposed.

This chapter answers the first question of the second part: After execution is free and knowledge is free, what valuable things are left in a person? Does it have a name, a structure, and is it trainable?

2. What’s left: naming scarce personal forms #

Start by double-checking the cleared shelves. "How much you know" is no longer valuable because the distance from "don't know" to "know", as mentioned in Chapter 1, has collapsed from "knowing the person" to "asking the question." "How fast" doesn't count anymore, that's what the zeroing curve is absorbing. Even "how fast you learn" is depreciating: the traditional definition of learning is loading knowledge into the brain, and the act of loading itself is what AI does best for you.

What remains on the shelf is the ability: faced with something that has never been done before, someone can quickly realize that "I have no usable basis for judgment", and then find someone who has done it before, take his approach together with the applicable conditions, judge which conditions do not hold true for him, change them, try it again at the minimum cost, and fix it if he makes a mistake. After a few rounds, this "never done thing" became a reusable standard for him. Others, faced with the same thing, start based on their feelings and work hard for ten years, leaving only busyness.

I call the previous ability adaptive insight. It consists of four actions, which are a cycle:

Discover standard gaps → Borrow external standards → Determine migration conditions → Quickly verify and correct → (Precipitate as your own standards)

Here we need to continue the corresponding relationship clearly stated in the preface: "the last scarcity" mentioned in the whole book is adaptive insight for individuals, judgment (next chapter) for organizations, and writing it down is the standard. So this chapter is not about some mysterious talent, but what scarcity itself looks like in an individual person. Seeing how it works in individuals will allow you to answer the larger question in Chapter 4: how it becomes an asset to the organization.

3. Why experience doesn’t count #

Hidden in the four actions is a judgment that will offend many people. Let me lay it out first: **Experience is an unverified standard. **

The raw material of experience and standards is the same thing, both come from what has been done. The only difference is one process: verification. This is how a standard is made: it clearly states the conditions under which it is established, it is tried in new scenarios, it is missed, it is revised, and its boundaries are drawn. A piece of experience is: it has been established, at least it seems to have been established, in a scene where the conditions cannot be remembered; then it directly obtained the right to explain.

"I do it all the time" really means "I don't die when I do it". The reason for not dying may be that you did it right, or it may be that the market is tolerant, your opponent is worse, you were lucky, or it may simply be that the price of your mistake has not yet been paid. Without verification, you will never know whether you have standards or survival. This is the mechanism why many people's abilities have not increased after ten years of work: they have diligently accumulated a lot of experience, but without the verification process, the experience will always stay in the status of "candidate standard": the quantity is increasing, but the quality is zero. What's even more troublesome is that unverified experience is not only useless, it also takes up trouble: the more a person believes that he has experience, the less likely he is to initiate those four actions when encountering new problems. Adaptive insight begins to sleep from the moment he has experience.This is the first old gene named in this book: empiricism, that is, replacing "verified" with "done", and replacing standards with qualifications. Its manifestation on individuals is to live ten years as ten replays of the first year; its manifestation on organizations has been seen in Chapter 2 (empirical authority), and Chapter 6 will give organizational-level replacement plans. This chapter deals only with the personal level: empiricism in yourself is more hidden than in your organization because no one is there to audit it.

4. Straight lines and bifurcations #

The value of adaptive insight can be drawn as a picture.

If there is a verified standard for something, the execution path is close to a straight line: follow it, verify it, and fine-tune it. From the starting point to the end point, every decision point along the way is narrowed down to one or two options by the standard in advance. If there is no standard for one thing, the path will bifurcate at each decision point: you choose A, take three steps and find something is wrong, go back and take B, and B branches into three more... The path you have taken covers the entire plane like a spider web. The two moves consume the same resources (your time and trial and error costs), but the trajectory lengths are orders of magnitude different.

This picture explains why among the four actions of adaptive insight, the first one is not "coming up with a solution", but finding the standard gap: realizing that you are standing in a bifurcation chart is the prerequisite for everything. It also leads to the personal version of the iron law: If there is no standard, stop implementing it. Pay attention to the true meaning of the word "stop" at the personal level: it is not to stop and wait, but to stop using the busyness of execution to cover up the absence of judgment, and to switch the action at hand from "moving forward" to "finding standards." The most dangerous thing about non-standard execution is that it feels so good: every step is moving, every day is busy, and busyness provides the illusion of growth, while the trajectory is spinning in place.

Five, the anatomy of four movements #

The four movements sound unpretentious, each with counter-intuitive details.

Action 1: Discover the standard gap. This is the most difficult of the four movements because it has to fight against your own body sense. There is no alarm sound when there is no standard execution; on the contrary, you feel fulfilled. There is only one reliable test method: ask the matter at hand, "What is the basis for my judgment at this moment, and where has it been verified?" If you can't answer it, the gap is there. Note that the answer "I have done something similar" does not count as passing, that is experience, and its quality was just mentioned in the previous section.

Action 2: Borrow external standards. After discovering a gap, the mediocre response is "Let me think about what to do", and the smart response is "Who has done it before?" The best solution to anything is to borrow, because in most cases the problems you face are not the first time humans have faced it, and groping is equivalent to paying the tuition paid by others again. This action has a fitting technical metaphor: when today's AI models encounter something they don't know when reasoning, the standard action is not to guess hard, but to call a search, borrow answers that humans have created, and then organize them into their own output. Machines have learned to borrow first and then answer, but people often feel that "borrowing" is degrading. There is an operational tip that is easily missed when borrowing: borrow the standard and its applicable boundaries. Just borrowing methods without borrowing conditions is equivalent to buying medicine without looking at contraindications. This foreshadowing will be expanded on in the next section using an expensive specimen.

Action 3: Determine migration conditions. Borrowed standards all carry implicit conditions of origin, and the implicit conditions are usually not written on the standard instructions, and the original creators themselves may not be aware of them. Therefore, the borrower must do a physical examination for it: This standard is established in the place of origin. What environmental parameters does it rely on? Which parameters have changed in my scenario? Are the changed parameters exactly the fulcrum on which it is established? If you fail to pass these three questions, the standard cannot be put online.

Action 4: Quickly verify correction. After judging the migration conditions, it only "seems to work". The last step is to let reality vote at the lowest cost: try it again on a small scale, in a short cycle, and at a low cost, so that mistakes can be exposed when they are cheap. Validation is the immune system of adaptive insight. The first three actions determine the quality of the item you borrowed, and the fourth action determines whether the wrong item can be stopped before it causes widespread infection. Execution zeroing also drives down the price of this process: trial and error has never been cheaper, which means the excuse of "too late to verify" has never been so tenable.

6. The first sample: verification occurs in another world #

The four actions form a cycle. If any link is broken, the previous accumulation will be invalidated as a whole. One person spent $100 million to demonstrate where and how to break.

The first half of Ron Johnson's resume is a genius narrative in the retail industry in any era: he first became famous for operating the "affordable fashion" strategy at Target, and later served as Apple's senior vice president of retail. Apple Store and Genius Bar are his works. In June 2011, JCPenney, a long-established American department store, announced that it would appoint him as CEO. Public opinion was almost unanimously optimistic: for someone who can make the Apple store the world's king of sales, saving a department store should be a dimensionality reduction blow.What Johnson did after taking office was to faithfully replicate his proven standards at Apple: good products with honest prices, no need for promotional tricks. He launched "Fair and Square" to offer low prices every day, abolished coupons, cut the annual sales from about 600 to about 100, and banned the words "sale" and "clearance" in advertisements. On February 1, 2012, this new set of standards was rolled out to all 1,100 stores at once. The usual practice in the retail industry is to pilot a product in a small number of stores for a few months. Such full-scale direct promotion is almost unheard of in the industry. Asked why he didn't test first, his answer was later quoted repeatedly: "We didn't test at Apple."

Results come quickly. JCPenney’s old customers are comparison buyers who have been fed by coupons for 30 years. Cutting coupons is part of their shopping ritual; when the coupons are gone, they will stop coming, and the young customer group Johnson wanted to attract did not appear as expected. In the first full fiscal year after the transformation, revenue fell by nearly 25% to only $12.98 billion, with a full-year loss of nearly $1 billion; in the last quarter, sales fell by about 30% year-on-year, with a single-quarter loss of $522 million. The board relieved him of his 17-month term in April 2013 and brought back his predecessor.

Using four movements to dissect this specimen, the conclusion will be much more accurate than the popular attribution of "arrogance". For actions one and two, Johnson gets full marks: he knows that the old model of JCPenney is dead (gap), and what he brings is not a slap on the head idea, but a set of standards that have been verified by Apple in billions of revenue (borrowed). The rupture occurs in action three: Apple's non-discount pricing can be established. The fulcrum is that customers buy exclusive products, there is no price comparison object, and price is trust; JCPenney's shelves are daily necessities with substitutes everywhere, and customers' trust is precisely based on the ritual feeling of "I cut a coupon." Every fulcrum of the origin standard does not exist in the new scenario. And action four was turned off by him himself. The full meaning of the sentence "We never test at Apple" is: this set of standards does not need to be tested at Apple because it has been verified at Apple. **It’s not that he has no standards, he has a set of standards that have been fully verified, but the verification occurs in another world. **

The most regrettable thing is the end: he actually retreated several times during his tenure (restored "sale" in advertising, announced the restoration of coupons at the end), but each time it was a passive retreat forced by plummeting sales figures. It happened after full rollout, and the price had been fully accounted for. If the same correction occurs in fifty pilot stores, it is tuition; if it occurs in 1,100 stores, it is an epitaph. Quickly verified and corrected by reality, the content can be exactly the same, the only difference is the price tag.

7. The second sample: myself #

Johnson broke in action three. To prove that this cycle is equally merciless to people who write books, I hand over the second sample to myself. The rupture occurred in action two, during the months of writing this book.

In October 2025, I launched a new round of financing for the company. The first judgment was whether to use FA (financial advisor). I spent almost no time: my previous entrepreneurial experience made me believe that FA was useless. This experience from the previous war directly gave me the right to explain. According to the definition in the third section of this chapter, its quality is clear at a glance: several contacts from another stage and another situation, without a control group and no applicable boundaries. This is typical experience, not the standard.

Under its command, I first connected investors and told the story myself, including making appointments with well-known partners of first-tier funds to chat face to face. Later I realized that those conversations took place at the wrong time and in the wrong setting (wrong setting, the same story is orders of magnitude less believable, which is the content of Chapter 8). The physical feeling is very fulfilling: I am meeting people every week and making progress every day; there is almost zero progress. I experienced all the busyness mentioned in the fourth section exactly.

In March 2026, the deadlock forced me to relax once and cooperated with two FAs: one was transformed from a local fund, and the other was owned by a US stock-listed company. There is basically no profit from the cooperation, it is purely my unilateral output. Please note what is wrong with this step: I took two samples at random without any selection criteria, and the bad results in turn "verified" my bias: you see, FA is really useless. Using unstandardized execution to test an experience will not result in verification but reinforcement of prejudice.

In May 2026, after being recommended by my peers, I contacted an FA that was highly praised by many entrepreneurs. Within a few weeks, I received a density of feedback that I had not received for more than half a year. Two things happened quickly. First, they helped me establish stratified standards for the capital market: what investors at different stages are verifying, what logic different types of institutions follow, and what kind of people should be met for what kind of project status. This set of standards was obviously not invented for me, it has always been there, just not in my context. Second, they worked with me to polish the business plan and roadshow, aligning the narrative with matching investors. Issues that I had not thought about for more than half a year were quickly clarified. The review only requires one sentence: when it comes to financing, finding the right FA is half the battle; and any peer who has raised funds can tell me this standard in thirty minutes.This is what the correct action should be. Action 1: Admit "I have never done financing at this stage, and I have no proven standards." Action 2: Ask "Who has done it?", research alumni in the same entrepreneurial community who have raised funds, and find an FA that has been verified by peers. Action 3 and 4: Cooperate and verify at a low cost, and change if it is not suitable. This cycle will be completed in a few weeks at most. I skipped it and replaced it with a piece of experience from the last war, with a price tag of over half a year. For a start-up company, that is not the cost of time, but half of the financing window.

Putting the two samples side by side, the fractures just complement each other: Johnson has a set of fully verified standards, but the verification happened in another world, and the break was in action three; I didn't even borrow it, so I turned off the "borrow" option using old experience, and then used two non-standard attempts to weld it to death, and the break was in action two. The person who wrote this chapter and the person who reads this chapter are on the same starting line: empiricism does not bully laymen, it only targets those who believe in their own experience the most.

8. Adaptive insight attitude: read Stay Hungry in its entirety #

After dissecting the two fractured samples, the winner does not intend to tell a business story, but a posture, because whether the four movements of adaptive insight can continue to rotate ultimately depends on the posture.

In 2005, at the end of his Stanford graduation speech, Steve Jobs delivered the words that were printed on countless posters: Stay Hungry, Stay Foolish. Posters usually translate it as "be hungry for knowledge, be humble and foolish" and regard it as an inspirational slogan. Put into the framework of this chapter, it is actually the dynamic description of the adaptive insight cycle, and the order of the two words should be read in reverse: **Foolish first: Only those who continue to admit their ignorance will keep the radar for finding gaps open; Hungry last: Only those who admit ignorance will keep borrowing ** from outside. On the other hand, a person who thinks he understands has his radar turned off and never borrows it again. His adaptive insight stops on the day he thinks he understands. Johnson's seventeen months at JCPenney coincided with the seventeenth month that Foolish closed. The most successful people at Apple have the hardest time admitting their ignorance in the New World.

Hidden in the same speech was a more concrete sample, much more solid than the slogan at the end. After Jobs dropped out of Reed College, he dropped in on a class that seemed useless at the time: a calligraphy class on serif fonts, word spacing, and the rules of great typography. When designing the first Macintosh ten years later, he moved all the standards from this course into the first personal computer with exquisite typography. His own review is: If he had not attended that course, Mac would not have multiple fonts and proportional kerning. Pay attention to the structure of this story: the borrowed standard (the layout rules of Western calligraphy) belongs to printing in its place of origin, and is transferred to a scene (personal computer) that is unimaginable in its place of origin. The reason why the transfer is established is that the fulcrum has not changed: the human eye's judgment of beauty does not change with the medium. This is a perfect demonstration of actions two and three: the value of borrowing does not depend on the origin of the standard, but on whether the fulcrum is still there. **

These two samples together answer the question "Is adaptive insight a talent?": Johnson does not lack talent, but attitude; in the story of the calligraphy class, there is no talent, only a person who keeps his radar on and puts it in the warehouse when he sees good standards. Adaptive insight is not a talent, it is four actions that can be trained. The first step in training is to change the default state of "I understand" to "I don't have a standard yet."

9. Zoom out: Why dare you call it an era? #

The title of the chapter uses the big word "era", so the reason must be given positively and cannot just rely on momentum.

The reason is: What is truly priced in each era is the most scarce ability at that time, and this scarce item is completing a change. In an era when knowledge is scarce, "knowing" itself is valuable: the imperial examination tests memorization, and the premium of a diploma is essentially a certificate of knowledge possession. The only literate person in a village can make a living by writing letters. In an era when execution is scarce, "doing it well and doing it fast" is valuable: the entire education and management system in the industrial age was designed around training people to become reliable executors. The organizational structures archaeologically examined in Chapter 1 are the fossils of this era. Of these two scarce items, one is being emptied by search and AI Q&A, and the other is being absorbed by the zeroing curve. Pricing power is handed over to whatever capacity remains on the shelf. This is not a market fluctuation in a certain industry, but a change that all people who work in knowledge and execution face, so it is worthy of the word "era".

This change of leadership has a special meaning for the principal leader, which deserves to be discussed separately. Looking at the four actions of adaptive insight from the individual to the company scale: The whole process of a startup company from 0 to 100 is essentially to find a bunch of standards that suit you. What standards should be used for recruitment, what standards should be used for pricing, what standards should be used for product iteration, and what standards should be used for spending money. Every step has been done by someone on the market, there are standards to borrow from, migration conditions need to be judged, and low-cost verification is required. The so-called strong entrepreneurial ability, if you break it down, this cycle turns quickly on the founder: others have been groping for three years, but he used a standard to bypass it in three weeks. The so-called company slows down as it grows up is often because this cycle is not passed on from the founder to the organization: only one person out of a thousand people is performing the four actions, and the remaining 999 people are executing without standards or empirically.This forces the question in the next chapter: no matter how strong a person's adaptive insight is, it is only that of one person. How did it become a thousand people?

10. Boundaries of Judgment #

There are three directions in which this conclusion can easily be pushed too far. Let’s stop one by one.

First, adaptive insight is not IQ or academic qualifications. None of the four actions relies on computing speed or knowledge stock: discovering gaps relies on honesty, borrowing relies on letting go, judgment transfer relies on sensitivity to conditions, and verification relies on being willing to let one's judgment be beaten. The correlation between these four things and intelligence is much lower than most people think. Johnson's IQ and resume are impeccable.

Second, this chapter does not belittle knowledge, let alone deep cultivation. What free knowledge changes is the premium of "possessing knowledge", not the usefulness of knowledge. Your knowledge stock determines the quality of your indexing when borrowing standards: knowing that this field exists, knowing who has done it, and understanding the terminology in other people's standards still rely on accumulation. Similarly, "ten years of standard accumulation" can only come from deep cultivation; what this chapter opposes is never staying in a field for ten years, but repeating the first year in a field ten times.

Third, truthfully explain the quality of personal samples. The loser sample in this chapter (Johnson) is a verifiable case at the financial reporting and reporting level; the calligraphy lesson on the winner side comes from Steve Jobs’s own public speeches and is a first-hand self-narration; my own financing sample is a self-narration and cannot be verified externally. You can listen to it at a discount, but it at least satisfies one property that other cases cannot: the person who wrote it has no motivation to beautify it. However, I have not yet produced a verifiable real sample of the "Two Decades" group of practitioners of the same age; it is a portrait that you and I have seen in this chapter, not a piece of evidence. To supplement this evidence, we need to use the same ruler to track real people for a long time, and this sample is being collected; until it is in place, please treat "two kinds of decades" as a strong hypothesis waiting to be verified, and Johnson's seventeen months is its best circumstantial evidence at present.

What to Do Monday Morning (No. 1 perspective) #

Two movements, fifteen minutes each:

First, do a four-question screening for yourself. Write down the three most important things at hand and answer each one: ① Is there a proven standard for this matter? ② If not, who has made it? Can the standard be borrowed? ③ What conditions have changed with the borrowed standard in my scenario? ④ What minimum-cost experiment can be used to verify it this week? The thing that cannot be answered by the four questions is the current boundary of your adaptive insight, and it is also the thing that you should pay close attention to personally.

Second, replace the most useless question in the interview. Replace "How many years of experience do you have" with: "Tell me about an example of a time when you imported a practice from elsewhere and modified it before you could use it - where the original practice was inappropriate and how did you find out?" This question directly tests the third most scarce of the four actions. For those who can answer the details, ten years is ten years; for those who can only say "judgment based on experience", you already know what ten years are for them.

Note (individual and team perspective): Use the same four questions for yourself, frequency is more important than depth: asking one thing once a week is better than doing deep reflection once a year. When implementing in the team, remember the rules in Chapter 2: The person who cannot answer the four questions is not unqualified, but because no one has given him this ruler.

Implementation of division of labor with Chapters 6 and 7 (to prevent duplication of self-examination) #

• This chapter only writes "what and why": the definition of adaptive insight, four actions, straight line/bifurcation metaphor, why experience is not the standard, all from a personal perspective ✓
• The stop rule only gives one sentence to the personal version, and the organizational process (who has the right to stop, eight-layer standards) is not expanded, leaving it to Chapter 6 ✓
• The borrowing standard only demonstrates "why borrowing is optimal" and personal operation essentials. The organizational action chain, bilateral samples, and four gates are not developed and are left to Chapter 7 ✓

Chapter Acceptance Self-Check (compare with the five acceptance standards of the chapter)1. The assertion can be restated in one sentence ✓, and is an inference of the core assertion (the core assertion asserts that adaptive insight is the last scarcity → This chapter gives the definition, structure and personal-level evidence of adaptive insight). #

2. Whiteboard frame diagram ✓ (straight line vs bifurcated spider diagram + four-action cycle).
3. External comparison and data ✓: Loser side Ron Johnson (financial report level data: -25% revenue, 17 months, single quarter -30%); winner side Steve Jobs calligraphy class + Stay Hungry gesture (first-hand text of Stanford speech); The author's financing samples have been written into Section 7 (2026-07-26, and the quality of the self-report has been stated at the conclusion boundary); the missing "Two Decades" age-matched control samples have been truthfully marked at the conclusion boundary (long-term collection).
4. 12 candidate sentences ✓.
5. "What to Do Monday Morning" Two actions from the perspective of position 1 + personal notes ✓.