Form 990–sourced benchmarks for the Denver metro, IRS §4958 reasonableness documentation, and compensation committee support — for boards that need a number they can defend.
Nonprofit executive compensation is not a smaller version of corporate executive compensation. It runs on a different evidence base, answers to a different regulator, and fails in a different way.
A Denver nonprofit board setting executive director or CEO pay is not managing say-on-pay risk. It is managing IRS §4958 exposure — the intermediate sanctions regime that can impose excise taxes on the executive who received an excess benefit and, separately, on the board members who knowingly approved it. The protection against that is procedural, and it has to be built before the vote, not reconstructed afterward.
There is a second layer in Colorado that boards often miss: nonprofits are employers, so the state’s pay transparency obligations apply to them on the same terms as anyone else. What you publish in a job posting and what your Form 990 discloses have to tell the same story. Our Colorado compliance page sets out the thresholds in detail.
The figures below are drawn from CauseComp, our nonprofit benchmarking platform, built from IRS Form 990 disclosures and adjusted for the Denver–Aurora–Centennial, CO metro area. They show total compensation — base plus bonus, other reportable compensation, retirement and deferred amounts, and nontaxable benefits — not base salary alone.
Median total compensation for an executive director or CEO at a $10M–$25M human services organisation in the Denver–Aurora–Centennial, CO metro:
$190,000
Form 990–sourced, drawn from IRS e-file disclosures. Rounded, and offered as a market read rather than a quotation.
That is one number, and one number is not §4958 documentation. The rebuttable presumption turns on appropriate comparability data — the percentile range, the sector and budget cuts, the package components, and the peer set behind the figure. A board that takes a median off a web page has a weaker record than it thinks. Those sit in CauseComp, with a report built for review.
See the full Colorado benchmarks →The gap that matters on this table is not between roles — it is between the 25th and 75th percentile within a single row. For a $10M–$25M human services organization in this market, that spread runs to six figures. Two boards can both land inside the market and be a six-figure distance apart, and both be defensible, because scope, tenure, sub-sector, and the executive who is actually available in that year all move the number.
The mistake we most often correct is a board that anchors on a single median figure with no record of why that percentile was the right one for their organization. That is not a compensation problem. It is a documentation problem, and it is the one §4958 punishes.
Colorado has no size threshold. A nonprofit with one Colorado employee is fully covered, which makes it the strictest state in the country for small organisations.
Every posting needs four things: the pay range, a description of bonuses and other compensation, a description of all benefits, and an application deadline. “Open until filled” is expressly insufficient — a real problem for nonprofits that leave hard-to-fill programme roles posted indefinitely.
The job opportunity notice is the provision that catches boards. Before making a selection decision, all Colorado employees must be told about the opening on the same calendar day, in writing, with time to apply. Then within 30 days of the new hire starting, colleagues who regularly work with them must be told who was selected and how to register interest in future roles.
For an executive director search run confidentially by a board committee, that pre-selection notice needs deliberate handling. The confidential-replacement exception is narrow — it covers replacing someone not yet aware of their separation, not a search the board simply prefers to keep quiet.
Full detail, including penalties and deadlines, sits on our Colorado compliance page.
Boards routinely benchmark base salary and stop. The §4958 reasonableness test applies to the whole package, and in this market the gap between the two is material.
Total compensation means base plus bonus, other reportable compensation, retirement and deferred amounts, and nontaxable benefits. Deferred compensation is the most common omission, because the accrual and the payment fall in different years and the benchmarking exercise tends to follow one or the other rather than both.
A board that benchmarks to base salary and believes it is at the market median for a Denver organisation of its size is looking at the wrong number — and it is the larger one an examiner tests.
Compensation paid to a disqualified person is presumed reasonable only where three conditions are met: advance approval by an authorized body without conflicts; reliance on appropriate comparability data obtained before the determination; and adequate documentation made concurrently with it. Miss one and the presumption does not apply.
“Appropriate” is where Denver boards most often fall short. A national median for “nonprofit CEO” is not appropriate data for an organisation in a tier 2 — high cost market — sector, budget size and metro all have to be in the cut, which is the entire reason the tables above are cut that way.
We have written the three conditions up in full, with the failure mode behind each, on our §4958 reasonableness review page.
Not every board decision needs a consultant. Many Denver nonprofits need defensible comparability data, a clean process, and a document for the minutes — nothing more. That is what CauseComp is for. Where the situation is harder — a founder transition, a merger, a deferred compensation arrangement, pay that has drifted outside the range, or a board already fielding questions — we advise directly.
For corporate and public-company work in this market, see our Denver executive compensation page.
It depends far more on budget size and sector than on the metro itself. We hold Form 990–sourced percentile benchmarks for Denver in CauseComp rather than publishing them here — a single figure without the comparability set behind it is not appropriate data under §4958.
It is the safe harbor for nonprofit executive compensation. Pay to a disqualified person is presumed reasonable if three conditions are met: the arrangement was approved in advance by an authorized body with no conflict of interest; that body obtained and relied on appropriate comparability data before deciding; and the body adequately and timely documented the basis for its determination concurrently with making it. If any one fails, the presumption does not apply and the IRS uses a facts-and-circumstances analysis instead.
Yes. Nonprofits are employers, and Colorado pay posting obligations apply on the same terms as they do to any other employer that crosses the relevant threshold. The practical consequence for a nonprofit is that published salary ranges and Form 990 disclosures have to tell a consistent story — they are read by the same donors, journalists and peer organizations.
Usually not. Appropriate comparability data has to reflect the organization actually being benchmarked — sector, budget size, and geography. The Denver-Aurora-Centennial metro sits in a tier 2 — high cost-of-labor tier, so national medians can materially misstate the local market in either direction. A board relying on an unadjusted national figure has a weaker record on the second condition of the rebuttable presumption than it thinks it does.
It depends on the decision. A routine annual review with a stable executive and a clean board process is often well served by a CauseComp subscription — benchmarks plus a board-ready report. A founder transition, a merger, a deferred compensation arrangement, pay that has drifted outside the range, or an organization that has already received questions warrants direct advisory support.
Yes. Form 990 Part VII requires reporting of compensation for officers, directors, trustees, key employees, and highest compensated employees, with Schedule J adding detail for higher-paid individuals. That disclosure is public, which is both why the data exists for benchmarking and why boards should assume their decisions will be read by donors, journalists, and peer organizations.
Reasonableness reviews, committee support, or a CauseComp subscription — tell us where the decision sits and we’ll point you to the right one.
Tell us what you’re working through and we’ll follow up directly.