CFO Pay Benchmarking Sources Compared: Filings, Surveys and Estimates

Choosing a CFO pay benchmarking source before the review is on the calendar

The compensation committee has moved your review up to March, and the only peer data you hold is a tab you built two proxy seasons ago. The tab has a short list of companies, no filing references, and a column headed "total" that you can no longer define. You open a browser and find five kinds of answer: a filing database, a survey sold to employers, an enterprise platform, a salary page with a single number on it, and a consulting firm's site that invites you to get in touch. Each describes itself as CFO pay data. They are not the same thing.

Most disagreements between sources come from what is measured, whom it covers and when it was collected, not from error on either side. A figure from a proxy filing and a figure from a self-reported survey can both be accurate and still be impossible to compare.

This article sorts the common CFO pay benchmarking sources by who they are sold to and what data they hold. It also covers the manual route, EDGAR plus a spreadsheet, and the parts of that work CFOCompensation's catalogue is built to do.

Four measures that all get called "CFO pay"

Before comparing sources, name the measure. A CFO pay figure usually falls into one of four types, and most confusion starts when two types are placed side by side.

SCT total. The Summary Compensation Table is required by Item 402(c) of Regulation S-K. It shows each named executive officer's compensation for each of the last three completed fiscal years, across eight compensation columns plus a total: salary, bonus, stock awards, option awards, non-equity incentive plan compensation, change in pension value and non-qualified deferred compensation earnings, all other compensation, and the total. Since the SEC's 2009 amendments, stock and option awards appear at aggregate grant-date fair value, not at the annual accounting charge. An SCT total is therefore a grant-date measure. It is not realized or take-home pay, and the pension and deferred-compensation column can move a total without any pay decision behind it.

Compensation actually paid. Item 402(v) pay-versus-performance disclosure reflects changes in the value of equity awards during the year, where the SCT reports grant-date value. It also covers the principal executive officer and an average of the other named executive officers. It does not give a separate CFO figure, so it cannot substitute for the CFO's own SCT row.

Survey-reported pay. Employer-contributed or self-reported figures, with definitions set by whoever designed the survey. A survey may report cash compensation, total compensation, or something between, and the label matters.

Estimates. Salary pages that publish figures that are self-reported, modelled or recruiter-sourced rather than tied to a filing.

Nonprofit CFOs add a fifth measure. Form 990 reportable compensation for employees is generally the greater of Form W-2 box 1 or box 5, and it covers the calendar year ending with or within the organization's tax year. It differs from an SCT total in what it counts and when, so the two should not share a chart without saying so. The IRS Form 990 instructions hold the exact definition.

One more precision matters for any filing-based work: proxy year and fiscal year are different things. A proxy filed in 2025 typically reports the fiscal year that ended in 2024, and off-calendar fiscal years break that simple one-year offset. Datarails, for example, describes its analysis as proxy years 2020–2025, covering fiscal years 2019–2024. A source that blurs the two can leave you comparing different years without realizing it.

How the main CFO pay benchmarking source types compare

The table below describes each source by buyer and data model, using what the providers visibly say about themselves. It contains no prices and no coverage counts. Check each provider's current materials before you rely on a description here, since products change.

Source	Typically sold to or accessed by	Data model
Manual EDGAR plus a spreadsheet	Anyone with time	Primary filings, keyed by hand
execsalaries.com	Free, no signup	Executive pay sourced from SEC DEF 14A filings, for a subset of public companies
Equilar	Issuers, committees, HR and advisers	Licensed enterprise data and surveys
Salary.com (Compdata / IPAS)	HR teams	Employer-sourced surveys
Pave	Employers	Employer-contributed compensation data
Pearl Meyer	Boards and management teams	Advisory service relationship
Datarails	Any reader of its published report	Vendor analysis of proxy filings; static report
Heidrick & Struggles	Any reader of its published survey	Self-reported survey of PE-backed CFOs
Comparably, Payscale, Built In, Robert Half	Job seekers, employers, recruiters	Self-reported, modelled or recruiter-sourced estimates

Filing-based sources. The manual route and execsalaries.com both start from DEF 14A filings. execsalaries.com is free without signup, presents executive pay sourced from those filings for a subset of public companies, and its homepage invites visitors to see how they rank. If you only need a quick look at disclosed pay for particular companies, a free filing-sourced site is a reasonable starting point. The questions to ask are how its set of companies was chosen, which years it shows, and whether each figure links to the filing.

Enterprise platforms. Equilar markets an Equilar Insight suite for executive-compensation benchmarking, incentive design and pay-versus-performance, sold as licensed product suites, with no published list prices. Its Top 50 Survey is participation-based, covering Russell 3000 named executive officer and Section 16 officer pay, and participation is free. These products are built for the issuer side of the table: committees, HR and advisers who benchmark many roles on a recurring basis. An individual CFO preparing for one conversation is a different buyer from the one these suites are designed around.

Employer surveys. Salary.com sells Compdata and IPAS surveys, including a U.S. executive survey, built on data that participating employers submit. Pave is a compensation-benchmarking platform for employers, built on employer-contributed data. The common feature is that employers supply the numbers, usually in exchange for access. The strength of a survey is a consistent definition across respondents. The limit is that the figures are not tied to a public filing you can open yourself.

Advisory firms. Pearl Meyer is a compensation consultancy that advises boards and management teams. That is a service relationship with the issuer, so the data comes with judgment and context. It is generally not something a CFO purchases in order to prepare for their own review, and this article does not suggest otherwise.

Published studies. Datarails, a finance software vendor, published "CFO and the C-Suite 2026", an analysis of proxy filings. Heidrick & Struggles, an executive search and leadership advisory firm, publishes a Private Equity-Backed CFO Compensation Survey. Both are static reports, free to read, and both are covered in the next section because their universes are narrow and the figures need to be quoted whole.

Salary estimate sites. Comparably, Payscale, Built In and Robert Half publish CFO salary estimates or guides. Their figures are self-reported, modelled or recruiter-sourced rather than filing-cited. They suit a rough sense of a job title across companies of every kind, and they answer a different question from the one a public-company CFO asks before a pay review.

Published CFO pay studies: read the universe before the number

When a search for an "average CFO salary" or a "CFO compensation study" returns a headline figure, the useful questions are which companies it covers, which measure it uses, and which year it describes. Four examples show why.

Datarails reported that median CFO total compensation, using the SCT Total column, rose from $2,388,694 in fiscal 2019 to $3,864,631 in fiscal 2024, an increase of 61.8%. The analysis covered 10,024 SEC proxy filings across 1,991 of the largest US public companies by Datarails' own selection, proxy years 2020–2025. That is a whole claim with its universe attached. It describes large companies, it uses a grant-date measure, and it should not be read as the median for CFOs generally.

Equilar's 2023 CFO pay trends report looked at S&P 500 CFOs and found median total reported compensation held at $4.7 million in both 2021 and 2022. Meridian Compensation Partners studied S&P 500 CFOs too, and reported median SCT total pay of $4.9 million in 2023 (proxies filed 2021–2024), with a 5.9% compound annual growth rate since 2020. These figures describe the S&P 500 under each firm's own method. They are not interchangeable with the Datarails figure, because the company sets, the years and the selection rules differ, and this article does not combine them into a single number.

For PE-backed CFOs, Heidrick & Struggles' 2025 survey, reported by CFO.com, found that US PE-backed CFO respondents reported average total cash compensation of $604,000 in 2025, up 5% year over year. It was fielded online in December 2025, and all data is self-reported. The measure is cash only, so it cannot be compared with SCT totals, which include grant-date equity. If you work at a PE-backed company, the closest you can come from public filings is a labelled stand-in: disclosed public-company peers, treated as an approximation and called one. CFOCompensation treats PE-backed pay that way, and private-company pay is not covered.

Broad labor-market statistics need the same care. The Bureau of Labor Statistics reports a median annual wage of $166,570 for the occupation "financial managers" in May 2025. That category covers far more roles than public-company CFOs, and it is not a CFO comparable. It is useful context on how wide the label "financial manager" runs, and nothing more for a pay review.

Coverage: whether your peers name a CFO at all

Even a filing-based source can leave a gap when the peer you want does not disclose the CFO. Who is disclosed depends on filer status, not on how large the company feels.

Under Item 402(a)(3) of Regulation S-K, named executive officers are everyone who served as principal executive officer, everyone who served as principal financial officer, and the three most highly compensated other executive officers. A non-SRC filer's CFO is therefore a required named executive officer regardless of pay rank. A smaller reporting company is different: its named executive officers are the principal executive officer and the two most highly compensated other executive officers, so the principal financial officer is not automatically included and its CFO may not appear in the SCT at all. Scaled disclosure also gives SRCs three named executive officers instead of five, and two years of SCT history instead of three.

Emerging growth companies add another layer. EGCs are exempt from the Dodd-Frank say-on-pay, say-on-frequency and say-on-parachute votes, and may use scaled executive compensation disclosure. They are also exempt from Item 402(v). A recent-IPO peer set can therefore contain companies that disclose no CFO row, or no pay-versus-performance data.

Release No. 33-11419 would reshape this picture if adopted. On May 19, 2026, the SEC proposed amendments to the public-company filer-status framework in Release No. 33-11419 (SEC fact sheet). As of October 30, 2026, the proposal has not been adopted, and nothing changes unless it is. According to Morrison Foerster, if adopted as proposed, about 81% of public companies would qualify as non-accelerated filers and become eligible for much reduced executive compensation disclosure. If the proposal were adopted, a CFO who is not among a scaled filer's principal executive officer plus two other named executive officers could drop out of that filer's SCT, thinning CFO peer coverage further. Check the SEC's website for its current status before relying on any of this.

The practical point is that coverage is a property of your peer set, and a source should show it. Ask whether a source tells you how many of the companies in a set actually disclose a CFO, or whether it simply presents a total. For help choosing peers by filer status and size, see the guide to executive compensation peer group selection and the explainer on named executive officers under Item 402.

Doing it with EDGAR and a spreadsheet, and where CFOCompensation fits

Manual EDGAR work is the real incumbent, and it works. It costs no cash, and a CFO with a spreadsheet and EDGAR can replicate the core of what any filing-based source does. EDGAR Full-Text Search covers filings submitted electronically since 2001, including exhibits. The SEC's fair-access policy caps automated EDGAR access at 10 requests per second, which matters only if you script your retrieval. The route is EDGAR company search, then the DEF 14A, then the Summary Compensation Table, then the principal financial officer's row.

The cost is time, and it comes in five tasks that a hand-built set has to handle on its own:

1. Identification. The CFO row has to be found by title. It may read "Chief Financial Officer", "EVP & CFO" or "Principal Financial Officer", and interim and former CFOs also appear.
2. Normalisation. SCT columns have to be lined up across filers and across proxy years.
3. Citation. A hand-built set rarely records which filing each figure came from, so you cannot trace it a year later.
4. Cross-checking. Nothing flags a keying error or a mis-read row.
5. Coverage awareness. Nothing warns you when SRC or EGC peers name no CFO.

CFOCompensation is built to do that assembly. The catalogue is available now. The CFO Pay Comparables Guide & Peer Workbook by Revenue Band is a method guide plus a workbook you fill in, and it contains no pay figures. The related catalogue titles cover the same method by sector, the Proxy Summary Compensation Table Reading Guide and Peer Build Worksheet, and the CFO Peer Set Builder and Disclosure-Coverage Worksheet.

The CFOCompensation app is not confirmed live, so what follows describes design, not a service you can use today. It is designed to let a CFO enter their own pay (base, bonus including non-equity incentive, equity grant value and other pay) and place it against peer CFO figures from DEF 14A Summary Compensation Tables, filtered by revenue band, sector and US region. Every peer figure is designed to carry a link to its source filing and a last-updated stamp. Former-CFO rows and flagged rows would be excluded by default. SCT CFO totals would be cross-checked against the Item 402(v) average non-PEO named-executive-officer total, with failures flagged and excluded. A disclosure-coverage indicator would show how many peers in the set actually disclose a CFO. To be told when the app is ready, join the waitlist. Pricing lives on the pricing page.

Percentiles in the app are designed to work like this: percentile means the share of peer CFO totals at or below your figure, calculated across the latest three available proxy years, and withheld when the peer set has fewer than five distinct employers. Here is a worked example with round, illustrative inputs. In a peer set of ten disclosed CFO SCT totals, a figure with seven peers at or below it sits at the 70th percentile under the at-or-below definition. For a longer treatment, see how a CFO pay percentile is calculated.

The limits are worth stating plainly. The 402(v) cross-check is a consistency rule, not an audit. Grant-date equity is not realized pay. A spreadsheet and EDGAR can reproduce the same comparison. Private-company pay is not covered, and the catalogue does not offer legal, tax or negotiation advice.

Matching a source to your trigger event

The right source depends on what brings you to the question. These pairings are starting points, not rules.

• An annual pay review at a non-SRC public company. Peer coverage is fullest here, because CFOs are required named executive officers. A filing-based comparison with citations gives you something you can trace if anyone asks where a figure came from.
• A stepping-up VP Finance or CFO-designate. Read the filings for the measure first. Meridian observed that pay for newly promoted CFOs tends to ramp up over two to three years, so the proxy year you choose changes what you see.
• An SRC, EGC or recent-IPO CFO. Check coverage before anything else. A source that cannot tell you how many peers name a CFO will leave the gap invisible.
• A PE-backed CFO. Use survey data and disclosed public peers, and label the public peers as a stand-in. The PE-backed CFO public-peer stand-in workbook is the catalogue title built for that.
• A nonprofit or hospital CFO. Use Form 990 data and keep it separate from SCT-based figures.

Whatever the source, organize the cited comparables you will bring to the conversation, and record the measure, the proxy year and the filing for each one. A pay-review file made of traceable figures holds up better than a single headline number, whoever produced it.

About these figures: CFOCompensation works from disclosed figures only, shown as filed and cited to the source filing; any third-party survey or study is named and labelled as such. Nothing in this article is legal, tax or negotiation advice. CFOCompensation and Rovaryn Digital Inc. are not affiliated with the SEC or the IRS. Private-company pay is not covered.