Average CEO Salary by Company Size: Reading the Bands

The Headline Number You Found Doesn't Match Your Board Packet

You searched "average CEO salary" the night before a comp committee call, and the number that came back sits nowhere near your reality — it describes a company ten or twenty times your revenue, with a board that has an Equilar subscription and a compensation consultant on retainer. That mismatch is not an error in your search. It is a structural feature of how CEO pay data gets reported: almost every widely cited average CEO salary by company size figure in circulation comes from the same narrow slice of the market — S&P 500 or Russell 3000 constituents, because those are the companies whose proxy disclosures are complete, structured, and easy for data vendors to aggregate. By the end of this piece, you will know exactly which disclosed band a given headline number actually describes, why the average and the median diverge sharply at the top of the market, and how to build your own band when the public record thins out below it.

What "Average" Actually Means in a Proxy Statement

Every company that files proxy statements with the SEC must disclose compensation for its top five most highly paid executive officers — the Named Executive Officers, or NEOs — in the Summary Compensation Table, per guidance from Meridian Compensation Partners (2025). That table is the raw material behind every average CEO salary by company size figure you will ever see. It is not a survey response or a self-reported estimate; it is a legally required disclosure, and since the SEC's Pay Versus Performance rule (Item 402(v), adopted August 25, 2022, effective for 2023 proxies) it must also be tagged in Inline XBRL — a structured format that is both human-readable and machine-readable, per SEC.gov (2024). That structure is precisely why S&P 500 and Russell 3000 pay gets aggregated into clean averages so easily, and why smaller, less closely tracked issuers do not show up in the same headline studies nearly as often.

The distinction between "average" and "median" matters more in executive pay than almost anywhere else in compensation data, because the distribution is heavily skewed by a small number of very large equity awards. For 2024, average S&P 500 CEO pay reached $18.9 million, according to AFL-CIO Executive Paywatch (2025) — well above the median S&P 500 CEO total compensation of $17.1 million for the same year, up 9.7% year over year, per the Equilar / Associated Press CEO Pay Study (2025). A separate calculation from the Harvard Law School Forum on Corporate Governance (2026) puts median S&P 500 CEO actual total direct compensation at approximately $17 million for fiscal 2024, driven chiefly by performance equity rather than base salary. Note what those three numbers already tell you: average and median for the exact same universe, in the exact same year, differ by roughly $1.8 million. If a single average CEO salary figure is doing the talking in your negotiation, ask which measure — and which index — it actually represents.

The Index You're Being Compared To

Not every large-company statistic describes the same universe. The Russell 3000 measures roughly 3,000 US companies and represents about 98% of investable US equity market cap, per LSEG / FTSE Russell (2025) — a far broader and less top-heavy population than the S&P 500's 500 largest constituents. That breadth shows up directly in pay structure: all non-CEO NEOs' total compensation equaled 38% of CEO pay in the Russell 3000 versus 31% in the S&P 500 for 2024, according to the Harvard Law School Forum on Corporate Governance (2025) — meaning CEO pay is proportionally more concentrated at the very largest companies. If you run a company that would sit inside the Russell 3000 but nowhere near the S&P 500, quoting an S&P 500 average CEO salary by company size figure in your own board discussion is comparing yourself to a smaller, differently structured peer set than the one you actually compete in for talent. A closer look at how pay by company size is disclosed across index tiers is worth its own read: /blog/ceo-pay-by-company-size.

Where the Public Record Thins Out

Here is the honest limit of this data, and it is the most important sentence in this article: the S&P 500 and Russell 3000 studies above describe roughly the top 3,000 US companies by market capitalization. Below that tier — the small-cap issuers, the privately held mid-market companies, the founder-led businesses doing tens of millions in revenue — comprehensive, aggregated CEO pay averages simply are not published with the same rigor, and any number claiming otherwise should be treated skeptically until you can trace it to a specific, named filing. We don't publish an invented average for a company at, say, roughly $50 million in revenue, because no verified aggregate exists at that granularity; what does exist is the individual company's own DEF 14A, if it is public, or its Form 990 Part VII if it is a nonprofit large enough to file one. A dedicated look at that band lives at /blog/ceo-salary-50-million-revenue-company, and the broader small-cap picture — including how thin peer coverage gets and why that matters for negotiation leverage — is covered at /blog/small-cap-ceo-compensation.

Owner-operators face an even starker gap. A founder or majority owner drawing a "salary" from a company they control is not participating in the same market that produces NEO disclosures at all — there is no comp committee setting that figure against external peers, and no proxy requirement forcing a public number. That is a different question from "what does the market pay a hired CEO of a company this size," and conflating the two produces bad negotiating math in either direction. See /blog/small-company-ceo-salary-owner for how to separate owner draw from market-rate CEO pay when you are pricing a transition, a buyout, or an external hire into that seat.

Internal Promotion, External Hire, and the Band You Actually Sit In

Company size is not the only variable that moves the number — how the CEO got the job matters too, and it is disclosed just as reliably. Externally hired S&P 500 CEOs earned median total pay of $10.99 million versus $7.76 million for internally promoted CEOs, a gap of roughly 41.6%, per Equilar (2015). More recent succession data shows the mix shifting: internal hires made up 59% of Russell 3000 CEO successions and 77% of S&P 500 successions in 2024, according to The Conference Board (2024). If you are negotiating an incoming offer as an external hire, or defending a renewal as an internal promotion, the relevant average CEO salary by company size comparison has to control for that variable too — an internal-promotion band and an external-hire band inside the same size tier are not interchangeable, and a comp committee that already has this data will notice if your comparison doesn't account for it.

Building the Band the Public Data Doesn't Give You

An average is a single number describing a distribution you never see. A percentile band — 25th, 50th, 75th, 90th — tells you where you actually sit inside it.

The fix for a misleading average is not a better average — it is a properly constructed peer set, sized and filtered by revenue band, sector, and ownership type, with disclosure sourced back to the individual filing behind each figure. For companies large enough to sit inside the S&P 500 or Russell 3000 disclosure universe, that peer set can be built directly from DEF 14A Summary Compensation Tables and the accompanying Pay Versus Performance XBRL data. For companies below that tier, or for nonprofit executive directors, the equivalent construction runs through Form 990 Part VII and Schedule J filings, and through your own sector's structural comparables where no single aggregate exists yet. A median CEO total compensation figure calculated specifically against the Russell 3000, rather than a generic average, is a meaningfully different — and more defensible — number to bring into a room: /blog/median-ceo-total-compensation-russell-3000 walks through that calculation in full.

If you need that band built and cited rather than assembled by hand from scattered filings, the Sector Benchmark Report constructs the peer set, sources every figure to its originating disclosure, and positions your own pay at the 25th, 50th, 75th, and 90th percentile of a group actually comparable to your company — not a headline average built for a business ten times your size. Review the report at /store/sector-benchmark-report before your next renewal or offer conversation, so the number you bring to the table names its own origin.