There is a thing that happens in small companies that nobody designs and nobody names and that is almost impossible to recreate once it is gone. Call it the room. Not the physical space, though the physical space is part of it. The room is the condition that exists when the people who understand the work are in the same place as the people making the decisions about the work. When the person who knows why the customer called yesterday is also the person who decides what to do about it today. When the knowledge and the authority live together instead of being separated by layers of reporting and process and intermediaries whose job is to translate between the two. In the room, problems get solved before they become visible. Not because the people in the room are smarter than the people in larger organisations. Because the distance between seeing the problem and being able to do something about it is short enough that the seeing and the doing can happen in the same conversation. The CEO had been in large organisations. Had seen what happens when that distance grows. Had watched problems travel up through layers of management, losing specificity at each level, arriving at the decision maker as a summary of a summary of a summary, the original texture and urgency replaced by a version that was accurate enough to be presentable and vague enough to be safe. Had watched the decision travel back down the same layers, gaining interpretation at each level, arriving at the people doing the work as an instruction that bore only a passing resemblance to what had actually been decided. This company did not have that problem yet. The CEO knew the word yet was doing a lot of work in that sentence.

Maya was explaining the green notebooks to a new person when the CEO first understood what the company actually had. The new person had started three weeks earlier. Young. Quick. The kind of person who asks questions that reveal they have already been thinking about the answer. Maya was walking through the notebooks not as an orientation exercise but as a conversation, pulling out specific entries and explaining not just what they said but why they had been written, what customer interaction had produced them, what had changed in the product or the process as a result. The CEO stood in the doorway and listened. What was happening in that conversation was not training in the conventional sense. It was not the transfer of information from a document to a person. It was the transfer of context. The new person was not just learning what to do. They were learning why things were done the way they were done. The history behind the current state. The decisions that had been made and the reasons for them and the things that had been tried and had not worked and the things that had worked better than anyone expected. This kind of knowledge does not live in a manual. It lives in the people who were there when the decisions were made. And it transfers only through conversation, through the specific kind of conversation where one person explains not just the what but the why and the other person asks the questions that reveal where the gaps are. Maya did this naturally. Without thinking about it as a knowledge transfer exercise or a culture building initiative or any of the other things organisations try to systematise later when the natural version has already been lost. She did it because she found the work genuinely interesting and because she liked the new person and because explaining something carefully to someone who was genuinely listening was one of the more satisfying things a working day could contain. The CEO made a note. Not about Maya specifically. About the condition. About the fact that this company had people who did this naturally and that this was not something that could be assumed to continue on its own as the company grew. The natural version works at twelve people. At fifty it needs support. At two hundred it needs a system. At five hundred it needs to be designed, intentionally, by someone who understands what is being preserved and why it matters. The CEO added that to the list of things to protect. Then walked through the rest of the office and found Owen.

Owen was not easy to find in the sense that he was not where the noise was. He managed a team of six people in the operational support function, the part of the company that made sure the product worked for the customers who were already using it, and he did this from a desk at the far end of the open floor that was slightly separated from the rest of the space by a structural pillar and two large filing cabinets that had accumulated there without anyone specifically placing them. He was sixty-one years old. He had the posture of someone who had been sitting at desks for decades and had made peace with this. His hair was grey at the temples and white above them. He wore a pale blue shirt with the sleeves rolled to the elbow and a pen tucked behind his left ear that he appeared to have forgotten was there. His team was at the desks around him. They were working with the quiet focus of people who had been shown how to do the work by someone who understood it and who had asked them to do it in a specific way for reasons that had been explained rather than simply required. The CEO watched Owen's team for a few minutes from a distance. Then walked over and introduced. Owen stood and shook hands. His handshake was firm and brief, the handshake of someone who considers the ritual complete when the ritual is complete and does not extend it for social reasons. "How is it going?" asked the CEO. "Well," said Owen. He said it without qualification, which was not how most people answered that question in the company's first year, most people saying well and then adding a but or an except or an although that revealed what well was currently costing them. Owen just said well. The CEO sat in the empty chair beside Owen's desk and asked him to walk through what the team was doing. Owen walked through it. Clearly and in sequence, starting with the overview and moving to the specifics in the order in which they mattered rather than the order in which they were easiest to explain. The CEO asked questions. Owen answered them with the patience of someone who has answered questions about their work many times and who has not become impatient with the questions because the questions still seem reasonable to them. Twenty minutes later the CEO stood to leave. "You should be in the room more often," said the CEO. Owen said nothing for a moment. "Which room?" said Owen. "The one where things are decided," said the CEO. Owen looked at his team. At the six people at the six desks, working in the way they had been working before the CEO arrived. "I am in the room that matters," said Owen. The CEO thought about this for the rest of the morning. Nora was at her desk when the CEO came through the support area at half past eleven. She was sixty-three. She had been with the company since what she called the second week, which was when the office above the dry cleaner had acquired its third desk and she had been the first person to sit at it. Not as a founder. As the first person outside the founding conversation who had heard what they were building and had said, without being asked and without negotiating, that she wanted to be part of it. She had been part of it since. She handled the customer relationships that required patience rather than just knowledge, the customers who needed to be understood before they needed to be helped, the customers who called not because they had a specific problem but because they had a general anxiety about whether the thing they had trusted with their data and their time was worthy of that trust. Nora was the person those customers talked to. And after they talked to her they were, almost without exception, quieter. Not because she resolved the anxiety with information. Because she made the customer feel that the anxiety had been heard by someone who understood why it was reasonable to have it. She was the best person the company had at this specific thing and she knew it and did not make anything of knowing it, which was its own kind of quality. The CEO stopped at her desk. "How are you?" asked the CEO. "Good," said Nora. She had a way of saying it that was specific rather than reflexive, the way of saying it that meant she had actually considered the question before answering it. "Is there anything you need?" asked the CEO. Nora considered this too. "Not right now," she said. "Ask me again in the spring. I have a check-up in March that I have been putting off and I expect it will produce a conversation worth having." The CEO filed this. Not in a document. In the specific mental cabinet where the CEO kept the things that would require attention before they required action.

Joel's systems were running six things simultaneously that nobody in the company knew were running. This was not secrecy. It was the natural invisibility of infrastructure that works. The things that work do not announce themselves. They simply operate, quietly, in the background, making everything else possible without drawing attention to the fact that they are doing so. The things that fail announce themselves loudly and immediately and produce exactly the kind of urgency that makes it impossible to think clearly about what to do next. Joel had designed his systems to fail quietly, with enough warning that the failure could be addressed before it became visible to anyone outside the immediate operational team. Not because he expected them to fail. Because he had seen what happened in organisations that designed for success without designing for failure, and the picture was not encouraging. He showed the CEO the systems on a Tuesday afternoon in the meeting room with the whiteboard and the four chairs. Not a presentation. A walk-through. Joel at the whiteboard, drawing as he talked, the kind of drawing that is thinking made visible rather than a finished product being demonstrated. The CEO asked questions. Joel answered them without the defensiveness that technical people sometimes bring to conversations with non-technical leaders, the slight tension that comes from not being sure whether the question is genuine curiosity or the beginning of a criticism. The questions were genuine. The CEO had not built these systems and did not fully understand the technical decisions behind them. But the CEO understood what they were for and why they mattered and what the company would look like without them. That understanding was enough to ask the right questions and to listen to the answers in a way that Joel found useful, the way that being genuinely heard is useful because it clarifies your own thinking by forcing you to explain it to someone who is not already inside it. By the end of the walk-through the CEO understood three things. First, the systems were good. Not perfect, Joel would not have claimed perfect, but good enough to support the next phase of growth without major rebuilding. That was rarer than it should have been. Second, the systems were underdocumented. Joel understood them completely. Nobody else did. If Joel left tomorrow the operational layer of the company would keep running for approximately three months before the gaps in understanding became gaps in capability. The CEO made a note about this too. Third, and most importantly, the systems had been built with the next phase in mind. Not the phase after that. Just the next one. Joel had not tried to anticipate every possible future. He had built for the immediate horizon while leaving the architecture open enough that the horizon after that would not require a complete rebuild to reach. This was, the CEO thought, exactly the right approach. And it was the approach that most people did not take, because most people built for the present and told themselves they would deal with the future when it arrived, by which point the architecture that should have been designed for extension had become the obstacle to it. The customers were the third thing the CEO studied in that first week. Not the metrics. Not the satisfaction scores or the retention numbers or the revenue per account or any of the other measurements that described the customer relationship from a distance. The actual customers. The conversations. The support tickets. The feedback forms that most companies collect and file and review on a quarterly basis if they review them at all. The CEO read everything from the previous six months. All of it. Every ticket. Every email. Every comment left at the end of a support interaction. It took three days. What emerged from those three days was a picture that no metric had captured and no report had described. The customers did not talk about the company the way customers usually talk about companies. They did not describe transactions. They described relationships. They named specific people, by first name, and talked about what those people had done and why it had mattered. They described moments where a problem had been solved not just adequately but thoughtfully, in a way that suggested the person solving it had actually understood the situation rather than applying a script to it. One customer had written a paragraph about a conversation with someone called Priya, who had apparently spent forty minutes on a call working through a problem that was only partly the company's responsibility to solve and had done so without once suggesting that the parts of the problem outside the company's scope were not her concern. The customer had written: I did not feel like I was talking to a support agent. I felt like I was talking to someone who actually cared whether my problem got solved. The CEO read that paragraph three times. Then went and found Priya. Priya was at her desk, headset on, in the middle of a call. The CEO waited. When the call ended Priya looked up with the slightly cautious expression of someone who is not sure why the CEO is standing next to their desk. "I read what a customer wrote about you," said the CEO. "I wanted to say thank you in person." "Which customer?" asked Priya. The CEO described the paragraph. Priya nodded slowly, the way people nod when they are remembering something specific rather than acknowledging something general. "That was a complicated one," said Priya. "The problem was partly ours and partly something on their end and partly just a thing that happens sometimes that nobody is really responsible for. I just stayed with it until it was sorted." "That is exactly what I wanted to thank you for," replied the CEO. "It is just what you do, isn't it," Priya responded. The CEO thought about that sentence for the rest of the day. It is just what you do. That was the culture. Not the values on the wall. Not the mission statement in the company handbook. Not the slides from the last all-hands meeting. The instinct, in the person closest to the customer, to stay with a problem until it was sorted because leaving it unsorted was simply not a thing she considered. That instinct had not been trained into Priya. It had been selected for, unconsciously, by a hiring process that valued the right things even before it knew it was valuing them. And it had been sustained by an environment where the people around Priya operated the same way and where the person at the top had spent a week reading support tickets and had then walked over to say thank you in person. Culture is not what you say it is. It is what you do when nobody is watching. And the CEO was watching now, carefully, and what the CEO saw was a company that had built something real without knowing exactly how it had built it. The job was to know. Because the things you do not understand you cannot protect. And the things you cannot protect you will lose, not through intention, but through the accumulated effect of a thousand small decisions made by people who did not know what they were trading away.

On the last day of the first week the CEO sat alone in the meeting room after everyone else had left. The whiteboard still had Joel's diagrams on it. Somewhere in the building a cleaner was moving through the corridors with a quiet efficiency that was its own kind of competence. Outside the windows the city was doing what cities do in the early evening, the particular hum of a million separate lives running in parallel, each one its own small system of decisions and consequences and things that worked and things that did not. The CEO thought about what had been found. A company that had built something it did not fully know it had built. Knowledge that lived in green notebooks and in the heads of people like Priya and in the systems Joel had constructed before anyone knew they were needed. Culture that expressed itself not in what people said but in what they did when the situation was complicated and the easiest thing would have been to do less. Owen in the room that mattered, which was not the room at the top. Nora saying ask me again in the spring. Hana and Tomás working late and not talking about it. The bottle of Spanish wine with the red and gold label that had been waiting for the right moment. It was fragile. Not because it was poorly built. Because fragility is the natural condition of things that grow. The bigger a company becomes the harder it is to maintain the conditions that made it good when it was small. The room gets bigger. The distance between the knowledge and the decision grows. The person at the top stops reading the support tickets because there are too many of them and because there are other things that seem more urgent and because that is not what CEOs are supposed to spend their time on. The CEO made a decision sitting alone in that room. Not about strategy. Not about product or market or funding or any of the things that would appear on the agenda of the next board meeting. A personal decision. About attention. About where to point it and how to keep it pointed there even as the pressure to point it elsewhere increased, as it would, as it always did. The decision was this: to keep reading the support tickets. To keep walking over to the desk of the person who had done something right and saying thank you in person. To keep asking Joel to walk through the systems even when there was no problem that required it. To keep sitting with Maya in the meeting room, not for any specific purpose, just to remain connected to what was being built and how and by whom. To remember to ask Owen which room he meant. To ask Nora again in the spring. It was a small decision. The kind that does not appear in any plan and does not produce any immediate visible result. It was also the most important decision the CEO made in that first week. Everything that followed, the growth, the pressure, the decision that would eventually put all of this at risk, would have been different if the CEO had held to it. The CEO would not hold to it. Not because the intention was wrong. Because the pressure was real and the time was finite and the things that demanded attention were loud and the things that needed attention were quiet, the way the most important things always are, right up until the moment they are not.

End of Chapter 2

Writer's Thought:

The CEO reads every support ticket in the first week. The question is how many weeks that will last.

Here is What is Broken. The CEO. The Room.

MarvinPro | March 2026

marvinpro.com

Think Simple.