At SHRM’s national conference this June, the head of the largest HR association in the country told a room full of HR professionals that their profession faces “extinction.” Johnny Taylor recounted asking 92 Fortune 500 CEOs whether they valued HR. Only 10 percent said yes. Thirty percent said the function delivered so little that their companies could live without it. He pointed to Bolt, which fired its entire HR team for “creating problems that didn’t exist,” and to Uber, which had just cut nearly a quarter of its people division.
It would be easy for HR to get defensive about that. I’d rather take it seriously — because the CEOs aren’t entirely wrong, and the reason they’re not wrong points straight at the fix.
The HR those executives don’t value is reactive HR: the version that shows up after the damage is done. It processes the resignation after the email lands, runs the investigation after the complaint, opens the job req after the team is already underwater. Taylor offered a sharper picture of what HR should be instead — a meteorologist. A weather service can’t stop the storm. Its entire value is telling you one is coming, with enough lead time to move the boats. That is the function worth paying for, and most companies don’t have it.
Reactive HR earns the verdict it gets
When HR only reacts, every conversation with the business is a bill for something that already broke: the strong performer who already signed elsewhere, the manager problem that already became a legal claim, the vacancy that already cost you a quarter of output. From the corner office, that looks like overhead that arrives with bad news and a process to slow you down. “Creating problems that didn’t exist” is unfair as a blanket charge, but it names a real feeling — HR that generates work without visibly protecting the business. Processing the paperwork isn’t worthless. It’s just not a strategy, and it’s the first line item a cost-focused executive starts cutting.
What forecasting actually looks like
A forecasting people function does for talent what a good CFO does for cash: it sees around corners and hands leadership a plan before the number moves. In practice, that means a few specific things.
- Flight risk before the resignation. Patterns in pay, tenure, manager, and engagement reliably flag who is likely to leave, often months ahead. Companies acting on this aren’t guessing — IBM cut turnover by about 30 percent using predictive models, and HP identified roughly 120 at-risk people and drove attrition in its pilot group to zero, part of an estimated $10 million in avoided cost.
- Hiring ahead of the plan. Headcount mapped to next year’s revenue, so roles open before the team is buried — not three months after.
- Risk before it’s a claim. Spotting the pay inequity, the manager pattern, or the policy gap while it’s still cheap to fix, instead of after it lands on a lawyer’s desk.
- Cost modeled, not guessed. Showing what the workforce will cost under different growth scenarios, so people decisions sit beside the financial ones instead of arriving as a surprise.
The math executives actually respond to
Put real numbers on it. Gallup estimates voluntary turnover costs U.S. employers roughly $1 trillion a year. SHRM puts the cost of replacing a single employee at 50 to 200 percent of their salary. And the Work Institute, after more than 120,000 exit interviews, finds that about 75 percent of those departures were preventable. Read those three together and the conclusion is hard to dodge: the most expensive workforce problem most companies have is one they could see coming and largely stop. A function that forecasts it isn’t overhead. It’s the early-warning system on the single largest controllable cost in the building.
Taylor told HR to prove “the ROI of every employee,” which can land cold. Flip it. The point isn’t to put a price tag on people — it’s to stop losing the good ones by surprise, and to spend the company’s money on talent the way a careful owner would: on purpose, with foresight.
How to tell which one you have
For an owner, the test is simple. Does your HR show up with forecasts or with paperwork? Four questions worth asking.
- When a key person resigns, is it a shock — or did someone flag the risk a quarter ago?
- Does HR bring you a hiring plan tied to next year’s revenue, or a job posting after the team is already stretched thin?
- When you ask what turnover will cost you next year, is there an answer with an actual number?
- Does HR arrive with problems for you to solve, or with risks it’s already managing?
If the honest answers point to paperwork, the problem isn’t that HR is worthless. It’s that it’s being used as a processor when it could be a forecaster.
The bottom line
The CEOs in that room weren’t rejecting HR. They were rejecting HR that only tells them what already went wrong. The function isn’t endangered because people stopped mattering — they matter more as they get more expensive and harder to replace. It’s facing a choice: keep mopping up the weather after it hits, or start forecasting it. The companies that win the next decade will have someone whose job is to see the storm coming. Make sure that someone is yours.
Sources
HR Dive — “SHRM CEO: HR faces ‘extinction’ and has ‘lost the plot’ on the future of work” (June 23, 2026)
Gallup — “This Fixable Problem Costs U.S. Businesses $1 Trillion” (voluntary turnover)
SHRM — cost of replacing an employee (50–200 percent of annual salary)
Work Institute — 2025 Retention Report, based on 120,000+ exit interviews (share of preventable turnover)
AIHR — predictive analytics in HR: case studies on attrition modeling at IBM and HP
Your HR Pro 