Research & Data
Private-sector hiring in the United States kept growing in late September but lost a little speed, according to ADP’s weekly employment estimate. In its NER Pulse released Oct. 13, ADP said private employers added an average of 24,750 jobs a week over the four weeks ending Sept. 26. ADP summed it up this way: “After five straight weeks of strengthening, hiring slowed.”
The dip was small. The four-week average for the period ending Sept. 19 now stands at 25,250 jobs a week, so the latest reading is lower by 500 a week. It is still the second-highest reading in the 12 weeks of history ADP published with the release, and well above the low points of the summer.
A revision moved the peak higher
The Sept. 19 figure is not the one ADP first reported. In its Oct. 6 release, ADP put the four weeks ending Sept. 19 at 23,750 jobs a week. The Oct. 13 release revised that up by 1,500, to 25,250. Earlier weeks were unchanged between the two releases: 22,500 for the period ending Sept. 12, 20,000 for Sept. 5, 16,750 for Aug. 29 and 12,250 for Aug. 22.
That revision matters for how the latest number reads. Measured against the figure first published for Sept. 19, the Sept. 26 estimate would have looked like another gain. Measured against the revised figure, it is the first decline since mid-August. ADP warns that “these numbers are preliminary and could change as new data is added,” so the Sept. 26 figure could itself be revised next week.
How far the gauge has come since summer
The longer pattern is one of recovery from a weak summer. The four-week average was 14,500 jobs a week for the period ending July 11, then slid to 8,250 for the period ending July 25, the lowest point in the published history. It stayed around 10,000 through mid-August before climbing steadily: 12,250 for the period ending Aug. 22, 16,750 for Aug. 29, 20,000 for Sept. 5, 22,500 for Sept. 12 and 25,250 for Sept. 19.
What the numbers measure
The NER Pulse is a four-week moving average of the weekly change in private employment, seasonally adjusted and built from ADP’s high-frequency payroll data. It is published with a two-week lag. ADP Research produces it with the Stanford Digital Economy Lab, and it comes out on Tuesdays except in weeks when ADP releases its monthly National Employment Report.
Because each reading averages four overlapping weeks, adjacent figures share most of their data. The readings should not be read as separate weekly job counts, and multiplying one by four does not give a reliable monthly forecast. The series also covers private employers only and is not the same as the Bureau of Labor Statistics payroll survey, which uses a different sample, methods and revision schedule.
Why it matters for job seekers
For people looking for work, the September readings point to private hiring that strengthened from its summer lull and was still adding jobs at the end of the month. One week’s easing does not show a turn toward job losses. It does mean the run of steady improvement has paused, at least on the preliminary numbers.
The catch
A 500-job difference in a weekly average is small, and the comparison depends on a figure that was itself revised by three times that amount a week later. The next NER Pulse, scheduled for Oct. 20, will show whether late September marked a plateau or just noise in a series that is still being revised.
Sources
- ADP, “ADP National Employment Report Preliminary Estimate for September 26, 2026,” October 13, 2026. https://mediacenter.adp.com/2026-10-13-ADP-National-Employment-Report-Preliminary-Estimate-for-September-26%2C-2026
- ADP, “ADP National Employment Report Preliminary Estimate for September 19, 2026,” October 6, 2026. https://mediacenter.adp.com/2026-10-06-ADP-National-Employment-Report-Preliminary-Estimate-for-September-19,-2026
Featured photo: ADP's headquarters in Roseland, New Jersey, in a photograph uploaded to Wikimedia Commons in June 2010. Photo: ADPDigital, via Wikimedia Commons (CC BY-SA 4.0). Cropped from the original.
