CAPA Is Not Shame── Turning mistakes into what the team learns, a rare kind of experience

Essay 36 of the Nichinichi Kore Kojitsu series. It opens with a colleague bowing his head over a deviation report, then digs into CAPA, the frame of our review work, through old and new words: shame and guilt, the five whys, a just culture, training on the ground, and the fable of the old man's horse. Stop measuring a mistake with the yardstick of shame, pick up a digging tool instead, and the wound turns into wisdom. A blueprint that follows that path.

01Before you look down and say "I did it again"

That day a colleague came to my desk holding a single sheet. It was a deviation report. He set it on the desk and, before I could say anything, bowed his head: "I'm sorry." His voice was small. He looked like a child who had come to be scolded.

What I wanted was not an apology. "What was the cause?" — that was all. But when shame comes first, a person admits their own fault before they get to the cause. And admitting is one thing; fixing is another.

Let me firm up the ground here first. Our work rests on an idea called CAPA. The word sounds hard, but the substance is like putting out a fire in the kitchen. Scorch the pot, and first you turn off the flame. Next you think about why it scorched. Was the heat too high, or did you look away? Then you change where things sit, or set a timer, so the same slip won't happen. That is all it is.

Let me walk one more step through the order. CAPA usually runs like this.

• Identify the problem. Make clear what happened, when, and where.
• Dig into the "why" underneath it: root cause analysis. Don't stop at the surface slip.
• Fix the gap that is happening now: corrective action.
• Take steps so the same cause never sprouts again: preventive action.
• Revise the procedure. Put the fix down on paper.
• Tell everyone involved about the changed procedure: training.
• A while later, pull a few cases and see whether it was really fixed: checking effectiveness.
• Don't let it end once; make it an ordinary habit: settling it into the culture.

This whole run is the backbone of GMP and GxP, the systems that protect a drug's quality. It is the backbone, so if it snaps, nothing stands. Before I talk about shame, I wanted to tread this ground firm.

02People always dress up a mistake

Reading through the reports, one sheet caught my eye with its oddly vague wording. "During work, an event was confirmed in which part of the record was not reflected." Who, in which step, dropped what? The heart of it slips right past. A passive sentence with no subject softly hides where the responsibility sits. Reading it, I felt like I was grasping at clouds.

Over this one sentence I laid an old line. In the Analects, left by the students of Confucius, there is a passage: "When the small man errs, he is sure to gloss it over." The small man, when he makes a mistake, always dresses it up. To gloss over means to trim the surface and hide it. Twenty-five hundred years ago, people were already doing the same thing. More than the mistake itself, they mind how they look for having made it.

Why do we dress it up? Let me borrow one clue from the anthropologist Ruth Benedict. She split the yardsticks that bind human conduct into two.

Under a shame culture, a mistake becomes something that must not be seen. That is why my colleague bowed his head before speaking of the cause. He is not weak. The yardstick of other people's eyes pushes him that way.

But here is the pitfall. The moment we feel ashamed and hide it, the learning that mistake carried is buried under the soil along with it. Why did the record drop out? Dug up, that answer would have become a treasure that guards the next time. As long as the yardstick of shame stays sitting on the desk, the cause sleeps on under the ground. So I want to stop the colleague as he starts to apologize and say it again: "You don't have to bow. Let's dig together."

03Five "whys," digging into the ground

One morning a missed check turned up. A phrase had slightly overstepped the range of the approved indication. It stayed in the material and went out the door. The junior who had handled it went pale. I did not want to make it a meeting to find "who is to blame." So I gathered us around the desk and opened with this: "The person is not the fault. Somewhere in the system there is a gap that lets things slip through. Let's look for it together."

The first answer came right away. Forgot to check. On the surface, correct. But stop there and the same thing happens again. We threw a "why" at it once more. Why did he forget? A last-minute swap came in just before the deadline, and the step of looking over the final version once more had dropped out. Then why did it drop out? Because it was never decided whose job the recheck-after-swap was. Why was it not decided? Because a rush swap was treated as an "exception" and was not written in the ordinary procedure.

This is the five whys. It is a tool systematized on Toyota's production floor, and usually about five digs bring you not to a person's carelessness but to a hole in the system. The number five is no magic. Think of it as a good measure for not being satisfied at the surface.

When the dug-out cause is not a single line, the cause-and-effect diagram works well. Put "missed check" at the tip of the spine, stretch thick bones from it — "people," "procedure," "tools," "how information travels" — and hang small bones with concrete causes on each. Think it through in your head alone and your eyes go only to the causes that come easily to mind. Draw the bones on paper and the branches you were overlooking come into view.

Here is a distinction I always tell younger staff. Chris Argyris, who studied how organizations learn, split learning into two. Single-loop learning and double-loop learning. A thermostat makes it easy. Turning the heat up or down to reach your set twenty degrees is single-loop. Asking "is twenty degrees even right?" is double-loop.

If we end this case with "I'll be careful," it stays single-loop. We went the way of doubting the premise. We abolished the "exception" of a rush swap, and rewrote the procedure so that for any swap, a different person looks over the final version once more. Root cause analysis is the doorway into this double loop. Fix only the surface, or doubt the premise too? At that fork, the work of digging makes us stand and choose.

04A gentleman's mistake is like an eclipse

When the digging settled, the junior who had reported it was still shrinking in his seat. I said, "Thank you for bringing it out and not hiding it." I meant it. If he had stayed quiet, that miss would not have surfaced for a good while. Because he raised his own hand, we could fill one hole in the system. I still remember his face as the tension left his shoulders.

At moments like this I always recall a line from the Analects. Confucius said, "To err and not to correct — this is called error." Everyone errs. What is asked is how you behave after you err. This one line quietly sheathes the blade of the one who would blame.

There is another, from the disciple Zigong: "The mistakes of a gentleman are like an eclipse of the sun or moon." An eclipse happens high in the sky, so everyone looks up and notices. There is no hiding it. But when the missing part returns to whole, people look up to that light again. A mistake is not to be hidden; rather, the act of setting it right is to be shown. Better to let people watch you fill back to full, openly, than to be ashamed of the missing part and try to cover it with cloud.

This is not just an old moral tale. The Harvard researcher Amy Edmondson found a common footing in teams that learn well and fix well, and named it psychological safety. In an unsafe workplace, people hide failures. A hidden failure stays as a hole in the system and one day becomes a large accident. In a place where you can speak up at ease, a small failure surfaces early and is fixed early. Confucius's words and today's research point the same way.

In medicine and aviation this is called a just culture. But it does not mean "anything goes." It does not overlook a wrongdoing born of cutting corners. The distinction runs like this.

This junior sits in the top row. So we don't blame him. Blame him, and the next person hides a failure. The "review team where you can speak up at ease" I wrote about in Essay 34 builds up this way, one case at a time. Dig out the mistake, offer it without hiding, fix it together. At the end of that repetition, like the waning moon filling again, the team grows one notch stronger.

05Fix it not on the paper, but in the doing

When I had printed the revised procedure and finished handing it around, I nearly called it done. I had pinned down the cause of the deviation, fixed the wording, and distributed it. As a process, it was complete. My manager's seal was stamped.

But on the way home that day it snagged on me. Will those people really read the pages I handed out? And if they read them, will they recall them in the middle of a busy review? I myself am doubtful how many of last year's revision documents I remember.

People make mistakes even after reading. This is not a story of laziness but of how the brain is built. Psychology has found that the range a person can hold attention on at once is very narrow. Write "be careful" in letters, and if it does not meet the eye at the moment the hands actually move, attention passes right by. So I changed my thinking: what should be fixed is not the wording on paper, but the doing. The making-visible I wrote about in Essay 35 works exactly here.

There was a Chinese thinker named Wang Yangming. He taught jijō maren. Sitting quietly and only reading books does not change a person; it is in real work and hardship that one is forged. He also said chikō gōitsu. If your head understands "this is how it should be" but your hands do not move, then you do not yet truly know, he says, coldly. Harsh words, but set against the review floor, they land.

I split the revision into two tiers. One was fixing the wording of the SOP itself. The other was embedding a device so that the procedure would move in the doing. These two have different roles.

I moved training from lecture toward practice. I brought in, as a physical object, a material where a deviation had once happened, and asked, "Try to find, with your hands, what went wrong here." A single moment of turning a page and finding it yourself lasts far longer than thirty minutes of just listening to an explanation. Chikō gōitsu is probably this: don't end by feeling you have understood; carry it to where the hands move.

A corrective action has not put down roots at the point of handing out. Only when someone's hands actually move on the floor, and stop or verify just as set, does it first take root in the soil. That revision, too — only when I saw the junior's finger pause once before the final version on the next case could I finally feel that this had come alive.

06Stack them one by one, and it becomes a culture

For a while after the revised procedure began to run, I watched for recurrence by sampling. Had the same kind of deviation shown up again in the next material? In none of the dozen or so I pulled was that mistake there. No flashy achievement. I was just quietly relieved. I let out one slow breath.

We call this work of "confirming that it has not shown up" checking effectiveness. Don't end at "I fixed it, I think"; let time pass, and back it up with facts. Only when the zeros continue can you say the correction worked. One success is not enough. An effect is an effect only when it lasts.

In Xunzi there is the phrase sekizen seitoku. Xunzi was a cool-eyed man who saw that, left alone, people drift toward the easy way. That is just why he did not count on inborn nobility, and bet on stacking small deeds. If the word "virtue" feels stiff, put it this way: repeat good deeds, and they become ordinary, until you can move that way without thinking. An organization is the same.

One CAPA is, on its own, small. Just one deviation in one material, fixed. But stack them, and the constitution of the organization changes. "We stumbled here before, let's be careful" becomes a watchword, and in time, without anyone saying it, the hands grow careful. This is settling it into the culture. A culture cannot be made by declaration. It roots only through stacking, one case at a time.

Here is one thing I want to confess to the younger staff. The role of carrying a CAPA is, honestly, hard. Sometimes your own oversight is the starting point, and shame trails along. But this role has a side that pays off. I call it the perk of the job.

So if you are handed a CAPA, don't lose heart. Before it is the cleanup after a lost point, it is the doorway to a rare experience. Cross swords with a department head, see through the wiring of the organization, lead people. Places that hand you such learning — and as work, no less — are not many. The spot you thought was shame has, before you noticed, become your best practice ground. Stack them one by one, and not only the thing you fixed but the self who did the stacking changes too.

07In all human affairs, the old man's horse

I remember that colleague from half a year ago well. When a deviation turned up in the explanatory material he had handled, he kept his eyes down the whole time in the meeting room. He said "I'm sorry" about three times. His face was red, and his fingertips moved a little on the desk. No one was blaming him, yet inside him, he was already being judged.

That same man, last week, was teaching a junior. He was explaining how to run a CAPA, drawing a diagram on the whiteboard. "Here, see, once you start looking for 'who is to blame,' you usually take the wrong road," he said. I heard it from the hallway and paused a little. In half a year, a person moves this far.

One old story. In the Chinese classic the Huainanzi there is the tale of the old man's horse. An old man living near the border loses his horse when it runs off. When the neighbors feel sorry for him, the old man says, "This may become good fortune." A while later the horse comes back, bringing a fine horse with it. Now everyone rejoices, but the old man says, "This may become misfortune." His son falls from that horse and breaks his leg. Then a war breaks out, and only the son with the hurt leg is not taken as a soldier, and survives. Misfortune and fortune cannot be told apart on the spot. Until you dig, you don't know which way it will turn.

A deviation, the misfortune, turns into learning, the fortune. I have watched that path all through this series. Move "where rightness sits" from inside people to inside the system. Treat a mistake not as material for judging but as material for fixing. Then, strangely, the quality of the materials goes up. What it took for that colleague to raise his lowered face was not talent. It was that he set the yardstick of shame down from his hand, just once.

Let me tie this to what I wrote in Essay 29. The story was that a compass does not teach the road. A compass tells you which way is north, but it does not tell you the road to your destination. The yardstick of shame is similar. No — it is worse. Shame throws even "north" off. The moment a person thinks they will be blamed, they start looking not for the cause but for an excuse. The direction itself twists inward.

So let me write the day a deviation was found in two ways. On the spot, it looks like a misfortune to be scolded for. You shrink, thinking "it's my fault," and out of shame you want it over quickly. But after you dig, another face comes out. It is a map that showed you where the weak point sits; you learn the cause was not you but a gap in the procedure; and the CAPA you ran has turned into wisdom you can hand to a junior. The same case, just by changing the yardstick you measure with, shifts its shape from misfortune to fortune.

So I recommend letting go of shame. This is not brazenness. It is not taking a mistake lightly, either. Just take the feeling of "embarrassing" off, for now, from the hand that digs the cause. Otherwise the shovel won't move. Grip the shame and a person keeps looking down at the edge of the hole forever. Only when you let go does the blade enter the ground.

And when you do dig, what comes out? That colleague showed it in half a year. The deviation, while buried, wore the face of misfortune. Dug up, it had become wisdom he could hand to a junior. A mistake, as long as you measure it with the yardstick of shame, is just a wound. Set the yardstick down, pick up the digging tool instead, and for the first time, the mistake turns to gold.