The Jobs Report Won't Tell Leaders What They Need to Know

When MillerCoors closed its brewery in Eden, North Carolina, more than 500 jobs disappeared. The loss was easy to count. What remained was not.

The region’s manufacturing knowledge did not disappear with the jobs. Neither did a 1.3-million-square-foot facility or the industrial infrastructure surrounding it. Nestlé Purina looked at the same loss and saw something else: productive capacity waiting for a new use. Purina turned the shuttered brewery into a pet-food factory, investing more than $450 million, repurposing 80% of the existing structure, and creating more than 300 positions.

That is what the jobs report will never tell leaders. It can count jobs created and jobs lost. It cannot tell a CEO what capabilities survived the loss, whether those capabilities can support the company’s next source of growth, or whether apparent economic weakness represents risk — or opportunity.

The First Draft Becomes the Verdict

On October 1st, at 8:30 Eastern, the Bureau of Labor Statistics will release the September jobs report. Economists will argue, markets will move and enormous confidence will gather around what is, by design, a first draft (and oftentimes a bad one).

BLS calls the initial payroll estimate preliminary and revises it twice as additional data become available. Revisions are not an occasional correction to the process; they are the process. And sometimes it turns out that the first estimate was not even directionally correct. On August 7, 2026, BLS cut its estimates for May and June by a combined 103,000 jobs. In November 2025, it had revised August payroll growth from a gain of 22,000 jobs to a loss of 4,000. The numbers did not just move; the story changed direction entirely.

And yet markets move on the first version anyway. BLS describes payroll employment as one of the earliest indicators of economic conditions each month and names financial markets among its major users. That still baffles me: The jobs report lands like a verdict while BLS is still taking testimony. Huh?

That uncertainty isn’t even the report’s biggest limitation. A perfectly measured national number still could not have identified Eden’s manufacturing knowledge or evaluated whether Purina could build its factory of the future there. 

The report is useful, I suppose, because it compresses a sprawling economy into information markets can trade. But compression removes detail. A number can be important enough to move billions of dollars and still be nearly useless for deciding where to place your next $450 million.

The problem for leaders is this: The macro story seeps into organizational assumptions. Talent is available. Hiring should be easier. Our turnover is normal. The headline becomes an explanation just when leaders should begin a diagnosis.

National Numbers Hide Local Reality

SHRM’s 2026 Talent Trends research exposes the disconnect. Amid a cooling labor market, 68% of the 2,094 HR professionals surveyed reported difficulty recruiting full-time employees, and 53% of that group said recruiting had become harder. Labor can appear more available in the aggregate while the people an organization needs remain stubbornly difficult to find.

ABM Industries operates as though there is no such thing as a single labor market. When I asked CHRO Raúl Valentín how an organization with approximately 113,000 employees recruits at that scale, his answer was simple: “The work we do is done locally. We’ve got to know the communities in which we’re recruiting.”

That sounds obvious until you consider how often “the labor market” is discussed as though a hospital in Phoenix, an office tower in Chicago and an airport in Atlanta are recruiting the same people under the same conditions. ABM does not need to know whether workers are broadly available. It needs to know whether the right people will apply for a particular role, at a particular client location, on a particular shift.

And even perfect local data cannot tell leaders whether the labor problem is outside the organization at all.

Benchmarks Aren’t Diagnoses

When Pete Stavros, KKR’s co-head of global private equity, joined me on The CEO Daily Brief, he described challenging leaders at companies where turnover approached 50%. Some answered that the rate was normal for their sector or simply how the economy worked. But “normal” is a benchmark, not an explanation. Stavros urged them to treat turnover as an operational defect: gather the data, identify the root causes, act, and keep measuring.

We saw the distinction in our work with the Georgia Department of Corrections, a story Joe Terry and I describe in our latest book Surrender to Lead. Officer turnover had reached 39%, and vacancy rates at some facilities exceeded 50%. Leaders saw several contributing forces, but exit data pointed most clearly to something no labor-market report could measure: how people felt they were being treated. Employees did not feel appreciated. 

As the department worked with us to improve their culture, turnover fell to 26% in just one year, and safety incidents at some facilities declined over 45%. Recruiting harder without changing the experience of employees would have created more activity around an incomplete diagnosis.

Leaders Need a Diagnosis, Not a Headline

These stories reveal the blind spots in both directions. In Eden, jobs disappeared while valuable capability remained. At GDC, jobs remained open while the employee experience pushed capability out. ABM operates in the space between them, tracking the market where hiring actually happens rather than where it is easiest to average.

The jobs report can inform expectations about demand, wages, talent competition, and the cost of capital. It cannot finish the analysis. Markets need a fast estimate. Leaders need a diagnosis at the level where a decision will be made.

So read the report. Watch the markets react to a number everyone knows will probably change. Then put it down and go find out what is true where you are.