[{"data":1,"prerenderedAt":18},["ShallowReactive",2],{"post-the-seat-closest-to-the-rules":3},{"slug":4,"title":5,"seoTitle":6,"excerpt":7,"content":8,"tags":9,"readingTime":15,"publishedAt":16,"published":17},"the-seat-closest-to-the-rules","HR is the seat closest to the rules, or: three people who sat in it","HR Is the Seat Closest to the Rules","I have worked with three people in human resources, and between them they landed almost exactly on the three points I would have drawn if somebody had asked me to sketch the range — one above what I expect, one well below it, one flat in the middle. What the exercise keeps telling me is not about the three of them. It is about the chair: the one seat that writes the rules, enforces them, and decides who else gets a seat.","\n## The observation\n\nI finished the last piece by saying that the other half of it was human resources, that the failures rhyme, and that the reach is what makes the second half worse. This is that half.\n\nAcross the jobs I have had, I have worked closely with three people in HR. Three is not a sample, and I would rather say that at the top than bury it at the end. What makes them worth writing down is how neatly they spread out. Without any arrangement on my part, the three of them landed on the three points I would have drawn if somebody had asked me to sketch the range — one clearly above what I expect from the role, one clearly below it, and one sitting flat in the middle. I have been turning them over for a couple of weeks, and the exercise keeps refusing to be about the three of them. It keeps being about the chair.\n\n## The one who set the standard\n\nThe first was on the other side of the table when I interviewed. Two things were visible inside the hour.\n\nThe questions were precise. They came from somebody who had worked out what the role actually required, rather than from somebody reading down a form that had been handed to them. And when the conversation moved into territory they could not judge — the technical half of what I had built, the part where an HR background gives you nothing — they neither bluffed nor waved it away. They passed it to the colleague beside them who could judge it, and then sat and listened to the answer.\n\nThat second habit is rarer than it should be. Deferring in front of a candidate costs a small amount of standing in the room and buys a better decision, and most people, in my experience, protect the standing.\n\nAfter I joined, the same person turned out to be the connective tissue of the place. Not by holding workshops about it. They arrived cheerful, they talked to people across departments who had no other reason to talk to each other, and they went home with colleagues rather than ahead of them. Underneath the cheerfulness was the thing I actually want to name: every person in the building was treated as a person, at every level, with no adjustment for rank.\n\nI would call that the baseline. It is what an HR function ought to manage on an ordinary Tuesday. And I still file them, honestly, as above average, which is the uncomfortable part of this whole piece. When the floor reads as an achievement, the thing being described is the distribution, not the person standing on it.\n\n## The one who served upward\n\nThe second I could read within a few weeks of their arrival, and the behaviour had a shape common enough that I suspect most people reading this have met it.\n\nServe the boss, and treat serving the boss as identical to serving the company. Attention pointed upward, calibration pointed upward, flattery pointed upward. Whether the people two rungs down liked the work, whether the middle managers had what they needed to do their jobs, whether anything in the office was quietly broken — none of it entered the frame. I do not think this came from any unkindness. It simply was not what was being optimised.\n\nThe tell is what happens to information travelling in the other direction. Where the HR function is aimed at one person, that person hears about a company that is fine. Everything that would complicate the picture stops one desk short, because the desk it would have to cross has decided its job is to keep the picture simple.\n\nI marked it privately as a failure rather than as below average, and I have not softened the judgment since. The function had been aimed at one person, and the entire reason the function exists is the other several dozen.\n\n## The same company, two more people\n\nThe third needs a phone call first.\n\nI approached that company at a point when it had no opening that matched me. I wrote to the boss, who passed my contact to the person handling recruitment. The third sentence after the call connected was that there was no suitable position, and that they were sorry.\n\nI have no objection to being told no. What I noticed was how fast it arrived and what it was made of. No question about what I had built, no attempt to work out whether an unusual candidate might be worth ten minutes of somebody senior, no curiosity of any kind. The answer was produced by matching me against a list of open requisitions, the list came back empty, and that was the whole procedure.\n\nThis is the argument from [[why-kpi-deserves-serious-skepticism|the KPI piece]] arriving in the place where it does the most damage. That seat is measured on filling the requisitions it is handed. It is not measured on whether the people it brings through the door make the company better in a year, because nobody has a way to attribute that, and anything nobody can attribute drifts out of the job. So the requisition becomes the work. A candidate who does not fit an existing box is not a lead worth developing; it is unpaid effort outside the scope, which is exactly where the question I described in [[admin-is-not-a-neutral-department|the admin piece]] gets asked: what do I get out of doing this.\n\nThe consequence is not subtle, and you can usually read it off the room. A recruiting function optimising for closed requisitions hires people who close cleanly, and that is a different population from the one the company needed. I ended up inside that company through a different door, which is its own quiet comment on the screen I had just failed.\n\nThe person I actually meant to write about there is somebody else. They kept good relations with colleagues, they were pleasant to deal with, and nothing on their watch went wrong. What was absent was the thing the first person had: the capacity to notice what somebody is carrying and to act on it before being asked. Average is not a gentle word in this seat. It means the chair is occupied and the reach that comes with it is going unused.\n\n## What the seat actually is\n\nHere is why three people have held my attention for a fortnight.\n\nI used to think of HR roughly the way I thought about admin — a low ceiling with a floor sitting just underneath it, a place where not much value gets made and not much gets lost. I gave my reasons for abandoning that view of admin last time: the reach is the entire company and the feedback belongs to nobody. Human resources has both of those properties, and one more stacked on top.\n\nHR is the closest thing a company has to a legislature. Attendance, leave, promotion, bands, who is hired, who is let go, what counts as a reasonable request and what counts as an unreasonable one — the rules are drafted there and enforced there, by the same people, often in the same week. Very little else in a company both writes the rule and rules on the rule.\n\nThat combination has a second consequence I would rather state plainly than imply. Discretion, money, and nobody watching is a standing invitation, and this function sits on all three at once: agencies and training vendors to choose, referral bonuses to approve, headcount to justify, offers to shape. I am not accusing anybody I have worked with of anything. I am saying the structure is built the way you would build it if you wanted that outcome, and structures shaped like that eventually find the person willing to take what is on offer.\n\n## A hiring seat reproduces itself\n\nThe strongest reason to care about who sits there is that hiring is close to the only decision that goes on to make more decisions.\n\nA person choosing candidates, with nothing pushing back, drifts toward candidates who resemble them. The version easiest to watch is age: an older person in that chair tends to find older candidates more legible, reads their pauses more generously, and the composition of the company moves a few degrees a year in that direction. I have seen the older version of this much more clearly than the younger one, and I would not push the symmetry harder than my evidence supports.\n\nThe mechanism is not prejudice in the loud sense of the word. Fit gets felt rather than reasoned, and what feels like fit is usually familiarity wearing a better word. An ordinary person in that seat will hire, over and over, a slightly weaker version of themselves — and each of those people, given a few years, will hire again.\n\n## The principle, stated tentatively\n\n> The HR seat does not do the company's work. It decides who does the work, and under which rules. That makes it the highest-leverage chair in the building, and in most buildings nobody is evaluating it.\n\nI want to be careful with that, because it reads as a complaint about three individuals and it is not one. The second person did what the structure rewarded. The recruiter on the phone completed the task they had been given. Neither structure was positioned to notice the difference, which is the part I keep coming back to.\n\n## What I have come to think, for the company I would like to run\n\nNone of this is settled, and some of it will not survive contact with a real payroll. But I have stopped treating human resources as a seat you fill, and started treating it as one of the two or three appointments that decide what everything downstream looks like.\n\nScreen it harder than the roles that look more senior. The two signals I would watch for are the ones the first person showed me inside an hour: whether they hand a question to somebody better placed to answer it, and whether they can describe an individual employee in specifics rather than in categories. Somebody who can only talk about staff in the plural is telling you something.\n\nSplit the requisition target away from the hiring judgment, so that no single person is measured on close rate alone. Seats filled is a proxy, and I have spent several pieces on what proxies do once they are left alone with the work they were standing in for.\n\nPut dull controls where discretion meets money. Vendor lists that outlive one person's phone, second signatures on agency contracts, referral bonuses with a paper trail. The [[why-dirty-work-deserves-serious-consideration|dirty work principle]] applies here the way it applies everywhere else: the record belongs to the system, the judgment belongs to a person, and the reason to build the record is precisely so that the person's judgment is the only thing left to evaluate.\n\nMeasure the state of the people rather than the count of the tickets. Walk the floor and ask three of them how the last month went. This is the same move I arrive at every time, and I notice I have now written it down in three consecutive pieces, which either means it is true or means I have found a comfortable sentence.\n\nAnd hire for empathy, out loud, as a stated criterion. There is no clean test for it, which is precisely why it has to be stated rather than scored. The alternative is to leave it out of the decision because it will not fit on a form, and leaving it out is how the seat gets filled the way it usually gets filled.\n\n## What I am not saying\n\nI am not saying HR is a cost centre or a department to be routed around. The argument runs the other way: it matters more than its job description admits, which is why a mediocre appointment there is expensive rather than neutral.\n\nI am not saying the first person was exceptional. What they did is the baseline, and the fact that it registers as exceptional is the finding.\n\nI am not saying the second was a bad person. They were aimed at the wrong target by a structure that rewarded being aimed there, though I will admit the explanation is more forgiving than I feel on some days.\n\nI am not saying three people constitute a distribution. They are three people. What generalises here, if anything does, is the shape of the seat rather than the size of the sample.\n\n## Where this connects\n\nThis is the second half of [[admin-is-not-a-neutral-department|the admin piece]], and the two are one argument seen from two desks. Both functions reach the whole company. Neither has anybody watching the outcome. Human resources adds the rule-making and the hiring on top, which is why I would now rank it as the more dangerous of the two.\n\nIt connects back to [[professional-matters-to-professional-people|professional matters to professional people]] as well, where the failure I noted was an HR function buying systems nobody wanted. I filed that under professionalism at the time. I would file it one level deeper now: a seat measured on visible activity will produce visible activity, and software that can be pointed at in a meeting is the most visible activity available.\n\nThere is a line to [[what-the-two-image-rule-misunderstands|the piece on the two-image rule]] too. The argument there was that counting is the cheapest way to cover an absence of trust. A requisition list is the same instrument aimed at strangers rather than at colleagues, and it fails the same way, by producing precisely what it counts.\n\n## What the literature already says\n\n**Henry Mintzberg, *The Structuring of Organizations*.** Mintzberg separates the technostructure — the analysts who standardise other people's work rather than doing the work — from the support staff arranged around the operating core. Human resources belongs to the first group more than the second, and that is the distinction I had been missing. The output is not a service. The output is the constraint everybody else operates under.\n\n**Lauren Rivera, *Pedigree*.** Rivera's fieldwork on elite hiring describes cultural matching: evaluators favour candidates who share their background and self-presentation, then report that preference back to themselves as merit. It gives the drift I described a mechanism, and a fairly unflattering one. Fit is felt first and justified afterwards.\n\n**Herbert Simon, *Administrative Behavior*.** Simon's argument is that organisations shape behaviour less by issuing decisions than by controlling the premises people decide from. Human resources supplies premises for a living. Whoever writes the premises has more effect on the outcome than most of the people making the decisions, and almost none of the visibility.\n\n## A postscript\n\nTwo pieces in a row about departments I had quietly written off, and it was the same error twice. I judged both functions by the ceiling of the value they could add and never once looked at where the floor sat. In both cases the floor turned out to be a long way below the ceiling, and in both cases the repair is cheap and the neglect is not.\n\nThe other thing I notice is that I have never told the first person any of this. They set a standard I have been measuring two other people against for years, and as far as they know they were only doing the job. That note is easier to write than the router was to buy, and I have not written it either.\n    ",[10,11,12,13,14],"Org Design","Management","Team Design","Process","KPI",10,"2026-09-02",true,1788356310713]