Two different measurements sit behind this site. Headcount benchmarks come from public professional profiles, classified job title by job title. Spend benchmarks come from companies’ own SEC filings. Neither is a survey, and nothing here depends on a company telling us anything.
Every company has a Finance team and an HR team. Almost none of them publish how big those teams are. So we count from the outside.
For each company we gather its public workforce signals: department tallies, and the real job titles behind them. No survey, no self-reporting, no login. A company never has to participate to be measured.
Each job title is read and placed in the ten-function SG&A map. Off-the-shelf department labels are wrong surprisingly often: about a third of labelled contacts move once you read the actual title. Procurement buyers filed under Finance. Whole Sourcing and Facilities teams hiding inside Operations.
The published level is the people we actually observed, capped at each company’s profile-visible workforce so stale records cannot inflate it. That makes every count a floor. It is why the share of SG&A leads on every page and the per-1,000-employees figure follows it.
Reported SG&A is not comparable across companies, because filers put different things in it. One company books distribution costs there; its competitor books them in cost of sales, and looks leaner for no operating reason at all.
So the spend pages use a normalized line: reported SG&A less the disclosed items that distort comparison, namely advertising, R&D, pension, stock-based compensation, restructuring and one-time items, plus logistics reclassed where a filer parks it in SG&A. What is left is the residual overhead line, on the same footing for every company in the peer set. Every adjustment traces to a number in the 10-K, and each company is compared against US-GAAP peers in its own sector rather than a broad average.
Where a company does not disclose a line, it is absent, not zero. That is why the advertising benchmark covers fewer companies than the SG&A one.
The paid census closes that gap: every person resolved from named public profiles, subsidiaries and off-platform population included, typically landing within 5 to 8 percent of true headcount. Start a benchmark or book a 20-minute intake.
Two sources. Headcount comes from public professional profiles and public company records. Spend comes from companies' own SEC filings, the 10-K and the 10-Q. Nothing comes from a survey, and no company has to take part.
No. A survey shows you what the companies that chose to answer chose to report. We measure every company the same way whether it wants us to or not. That is why you can see named outliers here instead of an average that hides them.
We count the people we can see in public profiles, so the count is a floor. A survey counts everyone the respondent reports. Compare on the share of SG&A, which holds at any size, and read the per-1,000 figure as a lower bound. The paid census closes the gap and lands within 5 to 8 percent of true headcount.
Every page carries its own as-of date, and the spend pages carry each company's fiscal year end. Briefings refresh as new filings land.
Yes. Add it on any benchmark page and we measure it the same way and email you the result, free. The paid census goes further: every person resolved, your own peer set, about a week.
SG&A Benchmarking
