Well, McKinsey just published a hundred pages on the future of HR. Five gaps to close. They include a fork in the road, which is to lead the transformation or get absorbed by IT. But here's the really strange part. Every one of those five gaps is literally the same gap wearing a different costume. And almost nobody reading the report is going to notice.
Thank you to everyone who invited me to keynote their collaboration week. It was a wonderful experience and I couldn't say nicer things about the people involved in that organization. And I also hope everyone had a reverent Father's Day. I know that I lost mine ten years ago, and every year on this day, it just reminds me how much I miss the man.
But as we get into the topic today, we're going to be taking apart the new McKinsey HR Monitor. Now, most reports like this give you a to-do list. Fix your workforce planning, fix your hiring, fix development, fix retention, fix your AI adoption. And they call those five different problems. Five work streams. I think it's primarily five places to put a budget line.
But I'm going to make the case that those are not five different problems. They are just one problem showing up in five rooms of the same house. And if you chase them as five, you're going to spend the next two years looking busy while the thing that actually determines your future gets decided by somebody else.
Now, this episode is for you if you've ever read a sharp piece of research about your own function and felt two things at once — that it was right, and that it was describing a building you don't quite live in. I hope by the time we end, you'll have a way to read this report and the next one like it that tells you where you should actually push. Ruthlessly pragmatic.
But before we get into it, I want to say a quick thank you. A shout out this week to Mark from Puyallup. I've known Mark for roughly 40 years and just want to say thank you for being a part of this community and supporting the work.
I'm a full-time content creator, so if you have an opportunity and you want to contribute, a couple of places you can do that. You can subscribe to my best-selling Substack, or right here on the podcast for $3 a month, you can go ahead and support us there as well. You don't get anything different other than the feeling of satisfaction knowing that you're helping someone who is trying to make a living bringing this kind of stuff to you every day. It's about three bucks a month. If you wanted to do it for 30, that works too. But there's a link here in the show notes.
And I also want to say thank you to everyone tuning in, whether you are listening from Orange, New Jersey or from Montreal, where I'm told we are now a top 100 management podcast. And to all of my Montreal listeners, thank you sincerely. Please go grab a double-double at Tim Hortons on me. Figuratively — you should buy your own Tim Hortons. Anyway, I appreciate you being here.
And look, speaking of getting that mandate defined, I want to tell you about something that I built for exactly this problem.
I'm going to call this thing the Mandate Protocol. And I built it because every single HR leader in my practice keeps asking the same thing — a structured way to sit down with their CEO or other business leader and get explicit on what the function is actually supposed to produce. Not the feel-good version, not the big gooey words like "strategic" version. I mean the real one: connectivity, decision rights, and measurement.
There are 15 modules delivered over five days. You can do the work at your leisure when you have space to think. And it's only $297. If you go through the whole thing and decide within 14 days it didn't deliver, just let me know, and I'm going to refund all of it, no questions asked. All of the risk is on me. I've had no one take me up on this offer yet. You can find it all at mytalentsherpa.com/the-mandate-protocol.
All right, let's get into this. Let me walk you through exactly what the report found, because I think the data is pretty good. This is McKinsey at its best when the diagnosis really hits.
Roughly 1,300 HR professionals, 5,500 employees, 10 different countries. Starting with workforce planning: 11%. Let me say that again. 11% of organizations are planning their workforce on a long-term capability-based horizon. Everybody else is doing headcount math for next quarter. And automation is rewriting which skills the business even needs, and the planning function is still counting heads against a budget. That's a problem.
Then development. 24% of employees reported zero training participation. None. And more than half are getting feedback once a year or never. So tell me again what our OD groups are really working on. And here's the detail that should stop you: HR leaders in the study overestimated both how much training people were getting and how much employees actually received development. So the function is misreading its own workforce on the very thing it's supposed to own, and they're taking way more credit than they deserve. That's a little scary, to be honest.
And then let's talk retention. Mobility is freezing up. People with good jobs are staying in them. Voluntary attrition is down. And when employees say why they stay, it's not the ones that have long-lasting hold power. The top answers are concrete. Pay, 52%. Work-life balance, 46%. Job security, 45%. Those aren't reasons connected to the function. Not connected to the impact they're having in the world. They're saying it's because the door's heavier and they want to be paid fairly while it's shut.
And then they talked about AI. The headline most people are going to probably repeat on this: adoption is creeping, zero to six points, depending on the domain. Most of it is still in pilots, most of it is parked in the administrative corners of the enterprise. The report's phrase for where this leaves the function is the one I think we should really sit with: a fork. HR either steps up to lead how humans and machines work together, or it is going to watch those responsibilities drift someplace else — probably into IT and finance.
And that's what the report says. Five findings, and the way it's written, you would file each one of those under a different leader on a different timeline, and probably in a different meeting. But here's where I want us to slow down a little bit, because there are a couple things underneath this report that the report itself doesn't really say out loud.
Here's the first one. Every single finding describes the same failure. See if you can get the pattern here. Workforce planning counts heads. Development counts attendance. AI gets counted as pilots launched. Same operating habit each time, five different times. The function measures activity and calls it performance. That's something we've been talking about for a while now.
The second thing is about who this report is really for. And I want to be careful here because obviously McKinsey does really good work. But if you read it closely, it is a dispatch from among the largest enterprises on earth. The prescriptions assume a function that already has a seat, is already operating within their mandate, already has the altitude to argue about agentic operating models. I think that's a real description of maybe 3% of the market. The other 97% is not deciding whether to lead the agentic future. They're still trying to get someone to define what their job really is. They don't have a strategy-talent gap. They have an altitude gap. They have a "nobody agreed what HR is for" gap.
And a report that opens with the agentic frontier does a lot of talking past the room that I think most HR leaders are actually sitting in. Not always, of course, but a lot of them. So you've got these brilliant people with genuinely good data describing the penthouse to a building where most of the teams are still trying to figure out who's in charge of mowing the lawn.
So let's name the thing underneath all of it. I think the reason workforce planning stays at headcount, and the reason development goes unmeasured, and the reason AI gets stuck in pilots, is not separate failures of execution. It's that the function is operating at the wrong altitude. And at that altitude, the only thing it knows how to do is measure activity and sentiment. Everything looking in the rearview mirror.
Think about what HR can actually see today in most companies. It can see how many people got trained. It can see the engagement score, probably aggregated — we've talked about that before. It can see requisitions filled, programs launched. What it actually can't see, in the language the business respects, is why any of that changed an outcome that the CEO or the board cares about. Throughput, margin, speed of a decision, risk that didn't materialize. Now that's the altitude problem. It's in some cases a visibility problem. The function is flying low enough that it can count motion, and it's too low to see whether the motion is producing anything at all.
So now pull AI back in, because this is where I think it gets sharp. And I wrote about it on the Substack a couple weeks ago. The core of this is very simple. AI dropped into an operating model that measures activity is just going to produce more activity. It's probably going to do it faster. But it does not say whether it's the right thing.
So, for example, if a recruiter's job is filling reqs, AI is going to help them fill those reqs quicker. And time to a productive hire doesn't necessarily move. If finance closed the books by hand, AI is going to help them close it faster. And the decisions required in the process are still waiting behind the same approvals. The tool gets better. The measurement stays the same. The behavior follows the measurement every single time.
So here's what AI actually is, in my view — and we talked about this last week at GHX. It's an X-ray. It makes the altitude problem visible. It shows you in real time how much of your work was actually organized around a task a machine can now do in seconds. And almost none of it was ever connected to an outcome. This is a work redesign moment.
And that's the part that reframes the whole report. That fork that McKinsey described — lead or get absorbed — I think it's real. But it isn't going to be decided by how sophisticated your AI is. It's going to be decided by one thing: whether HR can make human capital legible to the business before finance and IT build that accounting system first.
And the real important way of doing that is work redesign. Because look — someone's going to build a system that connects people to enterprise value. It's going to happen. The only open question is whose model it's going to live in. If finance builds it, talent's going to be a cost line in their spreadsheet. If HR builds it, the work design question — the question of what human judgment is actually worth in each pivotal role — it's going to stay right where it belongs, with the people team. And that, I think, is the fork. Become legible to the business or become a data source in somebody else's model.
So what do you actually do with this? Four plays.
Play one: take your single most pivotal role and redefine it by outcomes before you put any AI near it. Not the activities — the outcome. A recruiter doesn't fill reqs, a recruiter produces productive hires, measured by who's still performing 12 months down the road. Write that down. Because the moment you deploy a tool against an outcome instead of a task, you have changed what the system rewards. And I think that's the whole game in one move. And you can run it on just one role this week.
Play two: pick one number that connects a people decision to a business result and start carrying it into the rooms where money gets decided. Speed to productive hire, revenue retained in your pivotal roles, capability built against next year's strategy. One number in business language, every time you're in the room — share the thing no one else is going to share. That's how the function stops being measured at the vibe and sentiment level and starts being legible at the enterprise level.
Play three: use AI as a diagnostic before you use it as a tool. Deploy it as a thought partner. Instead of celebrating efficiency, read what it exposes. Where did better input still wait behind the same approval? Which roles turn out to be pure coordination? That's not going to be a technology finding. That's going to be a workforce architecture finding. And it's the most honest org diagnosis you're going to get all year.
Play four: if you're in the 97%, skip the agenda conversation entirely and go focus on getting your mandate defined first. A handful of sentences agreed in writing between you and your CEO on what your function is accountable to produce, what your decision rights are, how you're going to measure it, and the outcome that you actually own. The only bad answer when you get your mandate is if you don't know what it is. Even if it's not the mandate that you want, you're at least moving forward with clarity. Because you can't operate at an altitude or on a mandate that no one ever said. It's a guessing game. And this is not the time to be guessing.
So get your lease in your name before you redecorate the penthouse.
Now, I'm aware that I just took a hundred-page McKinsey report, agreed with most of it, and then told you the real report is somewhere the report barely mentions. And that's a lot to land on this Monday. But if you're sitting there thinking the homework is to go redefine a role, build a metric, renegotiate your own mandate — yeah, that's the homework. I didn't make the building. I'm just telling you which floor you're on. And if you want to go to a higher floor, that's what you've got to do.
And if you take one thing from today, let it be this: the future of HR isn't decided by how fast you adopt AI. It's going to be decided by whether you can make human capital legible to the business before somebody else does it for you. And that is how you're going to have humans above the loop versus ignored completely.
So thanks for spending time with me today. I appreciate you being a part of this community of senior leaders who want to rethink how human capital really works.
And if you're thinking about how to apply this in your own situation, let me point you to a couple of different resources. Propulsion AI is workforce intelligence for private equity. Their AI teammates surface workforce risk before the close. And equally important, and I think more powerfully, they help leadership teams drive execution afterwards. How do they do it? Well, they translate strategy into individual accountability, coach managers to define roles by outcomes, and give every employee a clear line of sight as to what actually matters. Good stuff, all through artificial intelligence. Learn more at getpropulsion.ai.
And if you're heading into the CHRO seat for the first time, or if you're close to it, that's exactly who I built all of this for. Find me over at mytalentsherpa.com. You can find the Mandate Protocol, you can find access to my private coaching, you can find access to the CHRO Chronicles — which is really, really good stuff. And of course, you can join the waiting list for the next CHRO Ascent Academy. And we're getting close to launching a couple of other training programs specifically around business acumen and readiness for AI. So I look forward to sharing some of those with you there as well.
And that's it for today. Until next time, keep raising the bar, keep making the work legible, and keep on climbing.
