For several days, I kept thinking about the review concerning the late arrival, although no longer because of the contradiction it had created. After everything we had examined, I knew there was little point in trying to decide from the outside whether that episode represented an exception, a particular condition or an operational failure without knowing more about the circumstances in which it had occurred. The review had served another purpose: it had shown, with considerable precision, where our knowledge ended and how far we could go with the information available.
Yet one question was becoming increasingly difficult to ignore.
Was everything we had been unable to discover really unknown to the hotel as well?
If I had worked there, I probably would not have begun by looking for another review or going through the website again. I would have spoken to reception. I would have asked how late arrivals were normally coordinated, what information they needed in advance, which situations tended to create difficulties and what had happened on that particular night. Perhaps afterwards I would have checked the reservation, the messages exchanged, the expected arrival time and who had been on duty.
Not because any of those answers were meant to prove the guest wrong; they might even have confirmed that something had failed. But they would have allowed us to observe a situation that, from the outside, appeared only as an outcome and that, within the hotel, was probably surrounded by details, decisions and context we could not see.
Once formulated, that difference seemed obvious, and yet it substantially changed the position from which we had been looking at the property. For several conversations, we had been trying to understand what someone who had never been there could discover. Now I was beginning to wonder how much of what we needed to know already existed inside the hotel and was simply unavailable to someone looking at it from the outside.
The late arrival was only one possible situation. I also thought about the rooms some guests described as particularly quiet and others that attracted comments about noise. From a distance, we could compare reviews, look for recurrence and try to recognise whether there were conditions capable of explaining that difference. Someone who worked at the property every day, by contrast, would probably know much more specific details. They did not need to have conducted an acoustic study to notice that certain rooms were further from a circulation area, that others experienced more movement at particular times, or that some might be better suited to someone expressing a particular need for rest.
Something similar could happen with breakfast. From the outside, we had found experiences involving adaptations and others in which a guest considered the variety limited. Inside the hotel, there was probably much more precise knowledge about which changes could be accommodated easily, which required advance notice and which requests went beyond what the operation was prepared to offer.
The more I thought about it, the clearer it became that much of the hotel experience depended on a kind of knowledge that rarely appeared in full on a webpage, in a policy or in a commercial description. It was not secret information, nor necessarily anything extraordinary, but something far more ordinary: what people had learned over time by working inside the property.
It could reside in the person who knew which room to offer someone asking for tranquillity; in the reservations agent who had learned which question to ask before confirming a particular category; in housekeeping, where someone knew a particular detail that would never appear on a commercial specification; or in the restaurant team, where someone could distinguish between an adaptation that could be made on the spot and one that required preparation. It could also remain with the person who remembered why a particular policy had been established and under what circumstances it made sense to interpret it with a degree of flexibility.
None of those people needed to think that they were managing the hotel’s identity. They were simply doing their jobs and, perhaps precisely because of that, this knowledge could remain so invisible.
For years, I had associated an organisation’s knowledge with what was documented: procedures, manuals, systems, reports, statistics or technical specifications. Now I was beginning to discover another layer, far less visible and much harder to organise. There were things a hotel knew because someone had learned them after years of welcoming different people, resolving unforeseen situations, observing behaviour and recognising patterns that had never been formalised.
That knowledge had not emerged from a meeting intended to define the property’s proposition, and yet it could explain better than many declarations what the hotel was actually capable of doing.
The idea brought me back to the formulation that had appeared several conversations earlier: first understand the person; then decide how to look after them. Until then, we had used it to recognise a common logic behind different gestures, decisions and memories. Now I was beginning to see it from another angle.
For that way of looking after someone to exist in practice, somebody needed to know things that were not always written down: what to ask, what to observe, what could be adapted, where a limit appeared, which previous experience was worth remembering and what response might be appropriate in a particular situation.
Hospitality could rest on a shared idea, but ultimately it happened through decisions, and every decision required some kind of knowledge.
That was when a distinction appeared that I had not considered carefully enough until then. I spoke about the hotel as though it were a single entity capable of knowing things. I said that it knew its guests, knew its strengths, understood particular circumstances or had learned from an experience. But when I tried to discover where that knowledge actually lived, the answer was considerably less orderly.
Part of it might be found in a system, another part in a policy, and some of it might remain in a reservation history, a note or something discussed during a meeting. At the same time, an amount that was difficult to measure probably continued to live only in the memories of particular people.
That realisation produced a new discomfort because it introduced a simple but important distinction:
it was one thing for someone inside the hotel to know something, and quite another for the hotel itself to know it.
I thought about an experienced receptionist who knew the differences between the rooms perfectly. Faced with a guest asking specifically for quiet, they might not need to consult anything; they would simply know which option was most appropriate. The decision might seem so natural that no one would even think of it as knowledge. It was simply part of what that person had learned to do.
But if someone new were at reception the following day, the room and the guest’s need would remain exactly the same. The only thing that would have changed was who stood on the other side of the desk and, with that change, something the hotel appeared to know could suddenly cease to be available.
That possibility offered a different way of thinking about some of the variations we had observed in the previous conversation. It did not explain any of them in particular, because we still did not know enough of the facts and there was no point in filling the gaps again with convenient hypotheses. But it allowed us to recognise a source of variability that until then we had treated mainly as an inevitable part of a human operation: two people could share the same intention to look after a guest well and yet not have access to the same knowledge when making a decision.
The difference mattered because a way of looking after guests could not be sustained by willingness alone. If certain behaviours depended on knowledge accumulated by only a few people, the property might provide extraordinarily consistent experiences while they were present and lose some of that capability when the shift changed, someone new joined the team or another department simply became involved.
I began to see the hotel as a network of distributed knowledge. Reservations might understand something reception did not yet know; reception might have learned something that had never reached those responsible for describing the property commercially; housekeeping might notice a particular feature of a room that appeared nowhere else; and the restaurant team might repeatedly encounter a need that the rest of the organisation still regarded as exceptional. Management, in turn, might know the reason behind a decision that, over time, the team continued to carry out without remembering why it had originally been made.
Each department might be seeing a different part of the same property. That did not necessarily mean that the hotel was disorganised. It probably happened, to a greater or lesser extent, in any operation where many people made decisions every day. What interested me now was something else: how much of that knowledge mattered to sustaining what the hotel said it was and, above all, which part of it genuinely needed to leave the place where it had been learned.
Because not all of it needed to become public information. A guest did not need to know the internal procedure used to organise a late arrival, but they might need to know that arriving outside the usual hours required prior coordination. Nor did they need to know how the team decided which room to allocate for a particular preference, although it might matter to them to understand that not every option offered exactly the same conditions. The same applied to a breakfast adaptation: the internal steps that made it possible belonged to the operation; knowing which needs could be accommodated and which needed to be communicated in advance could directly influence a choice.
The distinction was becoming more precise. There was knowledge required to sustain the operation, other knowledge that needed to circulate among those making decisions, and a part that needed to leave some trace outside so that someone could choose for better reasons.
Confusing those layers would be as problematic as ignoring them.
For a few minutes, I imagined how absurd it would be to try to solve the problem by putting the entire operation onto a website. No guest wanted to read a manual before booking, and no property could explain every exception, condition or particularity without making its communication impossible to navigate.
The solution could not be to show everything the hotel knew, but to recognise which part of that knowledge could influence a choice and which part could change the way we looked after someone who had already chosen.
What could change a person’s decision needed to be sufficiently understandable before they arrived. What could change the response to that person needed to be sufficiently available to those who had to make the decision inside the hotel.
That difference was beginning to bring order to a problem that, until then, had seemed much broader.
I returned to the contradictions we had observed. A discordant experience might point towards a condition we did not yet understand, but I could now see that it might also serve as a doorway into knowledge that had never been formulated. If a review mentioned noise, the question was no longer only whether the hotel could be considered quiet; it also mattered which room it had occurred in and whether those who worked there recognised any circumstance that could provide context. Similarly, when faced with a difficulty involving a late arrival, it was no longer enough to compare it with other experiences: we needed to understand what the operation knew about that situation that remained invisible from the outside.
Each contradiction could therefore force us closer to the actual workings of the property and reveal something the hotel had learned without yet turning it into explicit knowledge.
And that had a value extending well beyond correcting a description, because it was beginning to show that some of the most important knowledge could remain invisible precisely when things worked well.
When someone resolved a situation naturally, they rarely recorded everything they had needed to know in order to do so. When a team recognised certain needs, anticipated problems or routinely found answers, those capabilities could come to seem so normal that no one thought of them as something that should be preserved, shared or even named.
Sometimes the most valuable knowledge attracted no attention because its main manifestation was simply preventing something from going wrong.
But that naturalness came at a cost. If what enabled good decisions to be made continued to live only in particular people, it could also disappear when a shift changed, someone had a day off, went on holiday, a new person joined, a team member left, or the property grew enough to distribute decisions among more people. What had worked for years as an almost instinctive capability could suddenly reveal that it had never truly belonged to the organisation.
I found that conclusion particularly uncomfortable because hospitality had always depended on people capable of interpreting situations that no procedure could anticipate completely. Trying to document absolutely everything could produce the opposite of what we intended: hospitality reduced to predetermined responses might preserve information while simultaneously losing what made it human.
The opposite extreme did not seem reasonable either: relying entirely on individual experience and expecting knowledge to be transmitted simply by working alongside one another. Somewhere between those two possibilities, there had to be a way for the hotel to retain what its people had learned without trying to turn every future situation into an answer already written.
The question was therefore beginning to change. It was no longer simply how much the hotel knew, but how much of that knowledge would continue to exist when the person who had learned it was no longer there.
I thought about all the times an operation appears to function naturally because the people sustaining it have been doing so for years. There are teams that understand one another with few words, people who remember preferences without consulting a system and managers capable of resolving complex situations because they recognise circumstances they have seen before.
From the outside, that may look like culture, and it probably is. But a culture that can only be transmitted by working alongside one another also reaches its limits when the person who must interpret a situation changes.
If we wanted the logic we had found in that hotel—first understand the person; then decide how to look after them—to be more than the sum of the judgement of a few particularly experienced people, we needed to discover how it could remain available when those people were not there.
That did not mean getting everyone to respond in the same way. If the judgement consisted precisely in understanding the situation first, different responses could be perfectly consistent with one another.
What needed to remain was not necessarily the gesture, but whatever made it possible to understand why that gesture made sense.
That distinction was beginning to seem decisive. A procedure can preserve a response; an organisation also needs to preserve the reasons that make it possible to decide when that response is appropriate.
Until then, we had tried to understand the hotel by observing the signals it left. Now I was beginning to discover another layer behind many of them, far less visible: knowledge accumulated by people who had learned to interpret the property from within.
Perhaps the next problem was to ensure that this learning no longer depended exclusively on who happened to be on duty.
Management Perspective
An important part of what makes it possible to truly understand a hotel is not found in its descriptions, policies or systems, but in the knowledge accumulated by those who operate it every day. Reception, reservations, housekeeping, the restaurant team and management may know conditions, preferences, limits and possibilities that have never been explicitly formulated.
That knowledge is a strength because it enables better decisions, but it also becomes a fragile capability when it depends exclusively on the memory or experience of particular people. Managing it does not mean turning every situation into a procedure, but ensuring that what matters for understanding the guest and making coherent decisions can belong to the hotel and not only to whoever happens to know what to do.
The Case Study
In the hotel we have been observing, we had spent several conversations reconstructing from the outside circumstances that could probably be understood more precisely from within the operation. The experience involving a late arrival had left questions about advance notice, coordination, procedures and context; the differences surrounding quiet, adaptations or particular needs might also contain conditions that the public information did not allow us to observe.
The new problem was no longer simply to discover what the property knew, but to identify where that knowledge lived, who could use it and how much of it still depended on particular people rather than forming part of a shared understanding of the hotel.
A Question Before Moving On
If the people who know your hotel best were absent tomorrow, would the property still know how to make the decisions that seem natural today?