Small Business Hiring Plans Just Hit a Four-Year High. Payrolls Fell the Same Month.

Every NFIB Small Business Economic Trends report ships stamped across the top, in block capitals, for the newsrooms that get it early: EMBARGO 1 PM THURSDAY. July's copy carried a number under that header that small-business owners haven't posted since October 2022: a net 20% of them now plan to hire in the next three months.

Four days earlier, a different federal building had already published the opposite story. The Bureau of Labor Statistics reported that the country's actual payrolls went negative in July — down 23,000, with the two months before that revised down another 103,000 combined. This desk covered that report the day it landed.

Same month. Same country. One survey says owners want to hire more than they have in nearly four years. One government count says, on net, almost nobody is. Both are true, and the gap between them is the actual story — not a contradiction to referee, but a mechanism worth understanding before you decide what it means for your own hiring plan this fall.

The number underneath the number¶

Read past the headline "optimism index" figure and the July NFIB report is really a labor-market report wearing a sentiment survey's name. Of the ten components that make up the index, eight rose. The one that rose most was hiring plans, and NFIB says so directly: "Hiring plans improved substantially and contributed most to the rise in the Index."

The NFIB Small Business Employment Index — a separate, narrower measure that combines actual and planned employment changes into one number — rose 1.9 points in July to 102.1, up from 100.2 in June. That was its first increase after four straight months of decline, and it now sits above both the 2025 average (101.2) and the historical average (100.0).

None of this happened in isolation. Job openings owners couldn't fill hit 36% in July, up 4 points and the highest share since June 2025. Break that down and 31% have openings specifically for skilled workers (up 4 points), against 14% for unskilled labor (up 2 points). The gap between those two numbers is doing a lot of work: this isn't a shortage of warm bodies, it's a shortage of the specific people small operators need.

Why a survey of intent and a count of actual jobs can point opposite directions¶

These are two different instruments answering two different questions, and neither is wrong.

NFIB's Optimism Index is forward-looking and self-reported. It's built from a survey of roughly 887 member responses in July (a 10,000-firm sample, 8.9% response rate), asking small-business owners what they plan to do over the next three to six months. A hiring-plans reading of 20% doesn't mean 20% of small businesses added a worker in July. It means a net 20% said they intend to, later.

BLS's payroll number is backward-looking and administrative. It counts what actually happened to employer payrolls across the whole economy — every firm, from a two-person shop to a hyperscale data center operator — in the month that already occurred. Nothing in that number is a plan; all of it is a completed transaction.

A survey of intent and a count of completed transactions are always going to diverge some months. What makes July's gap worth a piece rather than a footnote is that NFIB's own commentary explains, in plain language, why the aggregate number can go negative while the small-business-specific one goes up:

That's the mechanism, stated by the people who run the survey: a large share of the capital driving GDP growth right now is going into construction and hardware that employs comparatively few people once built. If small businesses are where the next wave of actual hiring has to come from — because the AI-capex side of the economy structurally can't absorb it — then a small-business hiring-plans index spiking while national payrolls stall isn't a contradiction. It's what the handoff looks like before it's finished happening.

Look at the first two bars together. Hiring plans jumped the most of anything in the survey — and the complaint that owners can't find the workers they want jumped almost exactly as much, one month at a time, in lockstep. That's not two unrelated data points. That's the same tightening labor pool showing up twice, once as ambition and once as friction.

The turn: record plans are not the same thing as record hiring¶

Here's the part worth sitting with before you read "highest hiring plans since 2022" as a green light.

NFIB's own numbers show the gap between intent and action widened in July, not narrowed. Hiring plans rose 9 points. Actual hiring activity — the share of owners who reported hiring or trying to hire in July — fell 1 point from June, to 61%. And 51% of all owners reported few or no qualified applicants for the positions they were trying to fill, unchanged from June — that's 85% of the owners who were actually hiring or trying to hire. Thirty-two percent said "few" (up 5 points); nineteen percent said "none" (down 5 points, so the extreme end eased slightly, but the overall applicant-quality problem didn't).

Put plainly: more owners than in almost four years say they want to add headcount. The same month, fewer of them actually did, and the ones who tried mostly still couldn't fill the role. A record-high intention number sitting on top of a flat-to-declining action number is not confirmation that hiring is happening. It's a measurement of frustrated demand.

That matters for reading the rest of the picture, too. Consumer sentiment cratered to 51.0 in the same window — one of its sharpest drops in years, with only 8% of consumers now expecting their income to outpace inflation. Owners who are optimistic enough to plan a hire are doing so at the exact moment their own customers are telling researchers they expect to fall behind financially. Neither number is fake. They're measuring different people, at different points in the same economic cycle, and the honest read is that hiring plans this strong don't automatically survive contact with customers this nervous.

What to do with a plan you can't fully act on yet¶

1. 1

   Don't treat NFIB's headline number as permission — check your own demand first

   A national survey of intent is not a forecast for your business specifically. This desk has already flagged nine straight months of small-business transaction counts falling even as revenue held — run your own numbers before assuming national hiring optimism applies to you.

2. 2

   If you're hiring for a skilled role, budget more time and a stronger offer than you would have in 2024

   Openings for skilled workers rose 4 points to 31% in July — faster than unskilled openings. If the role you're filling requires a specific credential or experience, plan on a longer search and a wage that reflects the tighter pool, not last year's posting.

3. 3

   Get the paperwork ready before the offer, not after

   Going from zero employees to one — or adding a second — trips a specific, deadline-driven checklist: EIN and payroll tax registration, state unemployment insurance, workers' comp, I-9 and W-4. This desk's legal checklist for a first hire walks through it in order, which is faster than learning the order from a penalty notice.

4. 4

   Price the raise into your plan, not just the headcount

   A net 31% of owners already raised compensation in July, and 19% plan to raise it again in the next three months — both above their historical averages. If you're budgeting a new hire off last year's comparable wage, rework the math before you post the listing, not after an offer falls through on price.

5. 5

   Watch next month's actual-hiring number, not next month's plans number

   The plans figure already told you what owners want. The one that will tell you whether the labor market actually turned is whether the 61% "hiring or trying to hire" figure rises and the 51% "few or no qualified applicants" figure falls. That's the pair to track, not the headline index.

Frequently asked questions¶

The plan and the payroll aren't the same document¶

Here's what actually happened in July, stripped of the two competing headlines: small-business owners want to hire at a rate they haven't wanted to in nearly four years, and the labor market didn't hand them the people to do it, while the national count of actual jobs went the other direction entirely.

None of that is a mystery once you separate the instrument that measures intent from the instrument that measures outcomes. It's also not a reason to sit on your hands — a business with real hiring demand should still go look for the person, budget for the wage the market is actually asking, and get the paperwork lined up before the offer goes out. It's just a reason not to read "record-high hiring plans" as "the hiring already happened."

It's just business — and the plan is not the payroll.