You Can Define Your CFO in One Sentence. Try It With HR.

Ask a CEO what they expect from their CFO and the answer comes back in a sentence. Capital, allocated to its highest return. The COO? Capacity, allocated to produce more with the same or less. The CRO? Selling effort, allocated to grow revenue. Three seats, three clean answers, each naming a scarce resource and the growth it's allocated to drive. Then ask that same CEO what they expect from the person who runs the workforce, the one asset that touches every result the other three are measured on, and the answer changes shape. "Culture." "Talent." "Engagement." "Retention." The words come easily enough. But notice what they are. They're activities, things the function does. Nobody describes the workforce the way they describe capital, as something to be deployed where it produces the most return. That's the tell. It's the one seat where the CEO has an answer that sounds good and is built nothing like the others.

Finance has a denominator. One number every decision reduces to: capital allocated against return. That single idea is why the CFO seat is legible. Every activity inside finance ladders up to it, and because it does, nobody argues about what finance is for. Somehow HR is different. Where is the point that every people decision comes back to? It's difficult to reason about a function without a denominator. It's described by what it does, not the outcome it achieves.

But the workforce does have a denominator. It's just rarely been named with the same discipline. The allocation of human capability, against the outcomes the business needs to produce in order to grow. That's the mirror of capital allocated against return, and it's the question the workforce seat exists to answer. Do we have the right capability, in the right places, aimed at the results that matter? And if not, what do we move? Here's the objection, and it's fair. Capital is one clean number and human capability is not. You can't reduce a person to a figure on a balance sheet. True. But the harder something is to measure, the more it needs a clear organizing purpose, not less, because the thing you can't reduce to a number is the thing most likely to drift without one. Finance got its denominator handed to it by a century of regulation. The workforce never got that gift, so it has to be chosen.

None of this is new. The idea that the workforce is a strategic asset has been argued for thirty years. Every major consulting firm has published versions of it. Every wave of HR leadership has carried the banner. And after three decades, the seat is no more legible than when the argument started. The titles sound more strategic. The work doesn't. Which should tell you the problem was never a lack of vision or a lack of trying. The fix requires a few genuinely hard choices, and each one takes something away from the function as it exists today, not a rename, not a reorg, not a promotion to a better title. It means taking the function apart and distributing the pieces to the places where the work can be optimized. Renaming is easy. Everyone has done it. But the hard choices are exactly what the next generation of leaders is looking for. The stakes are higher now than they've ever been, and so is the payoff for the CEO willing to make them. What's different now is that the tools to actually make these choices finally exist. For thirty years the right moves were too expensive to execute. They aren't anymore.

And the pieces have been sitting in plain sight the whole time. In 1999, Pitney Bowes pulled the transactional work out of HR into self-service systems, and what was left could finally trace a line from a change in pay to turnover, and from turnover to profit. Patagonia refused to treat the workforce as a cost to be managed, decided it was central to the business, and ran the people function against that single purpose. Fossil took performance management for fifteen thousand employees off the paper forms HR had always owned and handed the judgment back to the managers who were in the room. Three companies, three different moves, three CEOs who saw that the workforce was where the advantage would be made. Each proved a piece works. None of them put all of the pieces together.

The major consulting firms have all published visions for the future of HR. Firms that charge millions per engagement, trusted by the most recognized brands in the world, staffed by some of the brightest people in business. The models aren't even wrong. The problem is they stop exactly where the work begins. They hand you a new design, a talent solution architect here, a people experience partner there, and leave you to pour your existing team into it. New theory, same people, same results. What they will not tell you, because it isn't a deliverable anyone wants to buy, is that the model changes nothing until you make hard choices about where the work goes and who is capable of doing it. They sell you the diagram. They don't make you confront the plumbing.

The structural move is the opposite of renaming. You don't make the function more strategic by adding strategic-sounding titles. You make it strategic by breaking the function apart and taking the pieces that are not strategic away. The transactional work leaves to agents. The compliance work leaves to legal and finance. The high-leverage work goes back to the managers who should always have owned it, supported by AI agents that have finally crossed the threshold where they can hold the load that once made the work too heavy to give back. What remains is core. It is critical to the business. And it is the only part that was ever truly strategic to begin with.

So what is that part, and how do you build it? The test for every role that remains is simple: it has to trace back to the denominator. If a role doesn't exist to allocate human capability against a business outcome, it's leftover, and it should have been subtracted with the rest. Run that test honestly and the function comes back not as a hierarchy of generalists and specialists but as a small system of distinct functions, each owning a different outcome, each built on a different education and experience, all pulling against the same denominator. That shared denominator holds the system together, not a reporting line. I've intentionally not given these functions titles, because the function is what matters, not what we call it. That's branding. And the table won't show the transactional work, because it isn't here anymore. What's named is only the judgment that remains once the processing is gone, which is why two of these functions don't even need a full-time seat.

Every function owns a different metric, and every metric traces back to the denominator. That's the proof it's a structure and not a set of new titles, because you cannot rename your way into a different metric. Each role demands a different person; the one who navigates conflict between an employee and a manager is not the one who decides what capability to buy and in what form. It's also why HR gets staffed the way it does. You'd never put someone in charge of finance just because they like numbers. Those seats demand a specific expertise. The people function never had one formally defined, so the bar fell to what was left, being good with people and willing to take it on. That's not a knock on anyone who filled the role. When you can't say what the outcome is, you can't say who's qualified, and the requirement quietly becomes temperament and availability.

One function deserves a closer look, because it answers the objection every CEO raises: people don't want a chatbot, they want a human. It's a real reaction pointed at the wrong thing. People don't resist technology because they prefer humans. They resist it because the systems they've been handed are bad, the phone tree that loops, the bot that can't understand the question. After enough of that, asking for a human is self-defense. Build a system that resolves the request instead of trapping them in it, and the reaction changes, because what people wanted was never the human. It was the outcome. Nobody asks for a teller to take out forty dollars anymore. Which is why one of these roles owns nothing but user experience and adoption, whether the systems are good enough that people actually want to use them. A capability nobody adopts was never allocated against an outcome at all. The denominator test again, applied to the work itself.

And here is why this whole set of choices can only belong to the CEO. Building systems that need fewer HR people is not a move most HR leaders are built to make. Not because they can't, but because of the missing denominator. A function with no agreed measure of its worth has only one way left to prove it: stay busy. Activity becomes the metric because no real one was ever named. So when you subtract from HR, you're not trimming the function, you're threatening the only evidence it has that it matters. Finance can shrink its team for efficiency, because its worth is tied to an outcome, not headcount. HR can't. Taking work out is the one move the function cannot make from the inside, because from the inside it reads as erasing the proof of its own value. It takes someone whose incentive is the result, not the size of the department. That someone is the CEO. It always was.

There's a harder consequence too. If each of these roles demands a particular education and experience, the people currently in the function may not be the ones who fill them. Not because they failed, but because they were hired against a bar that asked for temperament and availability, and the new roles ask for something else. The person who was good with people is not, by default, the one who can sit at the executive table and reason about human capability the way the CFO reasons about money. Different qualifications, and pretending otherwise is how you end up renaming without changing the work.

So here is where it leaves you, and it has to be you, as CEO, not your head of HR. You can keep buying the other version, the frameworks and strategic-sounding titles sold to you for millions, and wait the way the field has waited for thirty years for a transformation that never arrives that way. There's a tempting middle option too: hire a better HR leader and hand them the problem. It won't work. A better leader changes who runs the function, not what the function is. The denominator, the teardown, the decision about where the work belongs, those sit above the function, at the CEO's altitude. They have to be made before the head of HR is ever hired, and that person gets hired precisely because they're already aligned with them. You're not hiring someone to figure this out. You're hiring someone to execute what you've already decided.

So make the decision that is yours to make. Decide what your workforce is for, and give the function its denominator. Take HR apart and send the pieces where they belong. Build the sharp system that remains, designed around the outcomes you need rather than the activities HR has always performed. You already know how to do this, because you do it everywhere else. You can say in a sentence what you expect from your CFO, your COO, your CRO, and hold each to a number that matters. The workforce is the single seat where you have neither, not because it matters less, but because the discipline to think about it this way never existed until now. That's the gap. Not a missing program, not a missing hire. A missing operating vision for the one asset that touches every result the other three are measured on.

The companies that got pieces of this right did it because someone in the corner office decided the workforce was where the advantage would be won. They got fragments. The whole thing is still available, and still unclaimed, because the choices are hard and the tools to make them only just arrived. That is the opportunity. Not a better version of HR. The end of a question that has gone unanswered for thirty years, settled by the one person who could ever settle it. The work is yours. So is the advantage, if you take it.