The Unbenchmarked Roles: Two-Fifths of Reported Nonprofit Officer Pay Sits in Positions the Surveys Barely Cover

Ask what nonprofit compensation data exists and the answer is a familiar list: the chief executive, the finance chief, the development director, a few other chiefs. Published surveys concentrate there because survey economics demand it. The Form 990 corpus has no such constraint — organizations must report every officer, director, key employee and highest-paid employee — so it records something the surveys structurally cannot: how much of nonprofit officer pay sits in roles almost nobody prices.

The finding

Take the thirteen position families that dominate published survey instruments — the list is below, and it is deliberately generous. Everything else in the corpus is an “unbenchmarked role.” The unbenchmarked roles account for 30.9% of all officer filings (226,236 of 732,319) and 39.4% of all reported compensation — $64.6 billion of $164.2 billion across 2021–2025 filings.

• The commonly-surveyed thirteen: Executive Director / CEO; President; CFO / Head of Finance; COO; Chief Development Officer; Chief Human Resources; CIO / CTO; General Counsel; Chief Program Officer; Medical Director / CMO; Director of Development; Director of Finance; Chief Administrative Officer.
• Depth, not fringe: 54 roles in the corpus carry at least 1,000 filings each — and 41 of those 54 sit outside the surveyed thirteen.
• Concentration at the top is real but partial: the ED/CEO role alone is 37.3% of filings — which still leaves nearly two-thirds of officer filings, and a larger share of officer pay questions, below the top job.

The unbenchmarked share of dollars (39.4%) exceeds its share of filings (30.9%) — these are not the cheap seats. Clinical chiefs, deans, controllers, chiefs of staff, investment officers: the roles surveys skip skew toward the expensive specialized positions where a board has the least intuition and the most exposure.

Why it matters

The §4958 reasonableness framework does not exempt a role because no survey covers it — like services, like enterprises, like circumstances applies to the chief investment officer as fully as to the CEO. A board pricing an unbenchmarked role today typically extrapolates from an adjacent title or accepts the candidate’s number. The filings contain the actual answer; it has simply never been assembled into published form for most of these roles. That gap — two-fifths of the pay, a sliver of the published data — is, as far as we can determine, the largest blind spot in nonprofit compensation practice.

Pricing a role no survey covers

Where no published figure exists, the pricing conversation defaults to the two anchors available: the candidate’s current compensation, and the nearest adjacent title. Both are structurally biased — the first toward whatever market the candidate is leaving, the second toward a role that may share nothing with the one being filled but a word. The reasonableness standard does not soften for want of a survey line: like services, like enterprises, like circumstances applies to the chief investment officer and the general counsel exactly as it applies to the chief executive (Treas. Reg. §53.4958-4(b)). For the roles in this analysis the defensible sequence is unchanged — a cohort built from the filings, matched on size, mission and market — and the record is deep enough to support it: 41 of these roles carry more than a thousand filings each. The blind spot is in the published literature, not in the underlying record.

Method and scope

Scope of every figure on this page: 732,319 Form 990 filings, tax years 2021–2025 (2023 is the last complete filing year; 2024 and 2025 are still filling), covering the 107 curated officer roles CauseComp publishes, filings with a known NTEE sector only. No cohort statistic below 20 filings is reported anywhere.

1. (a) The 13-role surveyed set is CauseComp's own categorization of the position families that dominate published survey instruments; the boundary is stated in full above precisely so it can be argued with.
2. (b) Dollar figures are sums of reported total compensation as filed, 2021–2025, unadjusted; the span includes two incomplete filing years, so the sums describe the corpus, not any single year's economy.
3. (c) “Rarely surveyed” describes published position lists generally, not any specific vendor's product.
4. (d) Share-of-pay is not share-of-need: some unbenchmarked roles are priced adequately by adjacent labor markets; the finding is about the absence of published nonprofit-specific data, not about every such decision being wrong.

Frequently asked questions