Fractional CMO Cost and Pricing

The Employer Load, Corrected: What a Full-Time CMO Actually Costs in 2026

Short answer: almost every fractional-CMO cost page on the internet, including this one until today, understates the employer load by using the wrong denominator. The Bureau of Labor Statistics reports that benefits are 30.1% of total compensation in private industry. That is not a 30% markup on salary. Benefits are 30.1% of a number that already contains them, so as a markup on wages the same BLS data gives 43.0%. On a $225,908 salary the difference between those two readings is roughly $29,000 a year.

This section shows the arithmetic, separates the part that is exactly calculable from the part that is only a benchmark, and then makes a point that cuts against the usual sales pitch: the employer tax load falls as salary rises, because the largest legally required component is capped. A flat "add 30%" rule is wrong for a CMO in both directions at once.

1. The denominator, which is where the error happens

The BLS series people cite is Employer Costs for Employee Compensation (ECEC). It reports what an hour of work costs an employer, split into wages and benefits, and it expresses benefits as a share of total compensation. To turn that into "what do I add to a salary", you have to divide by the wage line, not by the total.

Source: U.S. Bureau of Labor Statistics, Employer Costs for Employee Compensation, first quarter 2026, series CMU1010000000000D / CMU1020000000000D / CMU1030000000000D (civilian) and CMU2010000000000D / CMU2020000000000D / CMU2030000000000D (private industry), retrieved from the BLS public API on 9 September 2026. The final row is arithmetic on the two rows above it: benefits divided by wages.

The two middle rows are BLS. The bottom row is division. If a cost page quotes "about 30%" and then adds 30% to a salary figure, it has quietly moved between those two rows without doing the division.

2. The part that is exactly calculable: 2026 statutory employer taxes

Benefits are a benchmark. Payroll taxes are not - they are set in statute, and for 2026 they are published in IRS Publication 15 (Circular E). There are four line items, and one of them is a trap that cost pages get wrong in the employer's favour.

Sources: IRS Publication 15 (2026), Circular E, Employer's Tax Guide - "the social security wage base limit is $184,500", "The Medicare tax rate is 1.45% each for the employee and employer", "There is no employer share of Additional Medicare Tax", and "For 2026, the FUTA tax rate is 6.0% ... If you're entitled to the maximum 5.4% credit, the FUTA tax rate after credit is 0.6%". IRS Topic no. 560 and Topic no. 759 carry the same figures.

The Additional Medicare Tax line matters at CMO salaries specifically, because it only exists above $200,000 and a CMO is usually above it. It is the one payroll line where crossing a threshold costs the employee and not the employer. Any cost model that treats it as a 0.9% employer cost on the excess is overstating the hire by several hundred dollars a year.

3. The load falls as the salary rises

Because Social Security stops at $184,500 and Medicare does not, the employer's statutory tax rate is a declining curve, not a constant. This is the single most commonly missed fact in fractional-versus-full-time cost comparisons, and it works against the fractional pitch: the more senior the hire, the smaller the payroll-tax penalty as a percentage.

Calculated from the 2026 rates in Table 5. Federal only: state unemployment insurance, state disability where it applies, and workers' compensation premiums sit on top and vary by state and by class code. FUTA is shown at the 0.6% post-credit rate, which assumes state unemployment tax was paid in full and on time and that the state is not a credit-reduction state.

At $225,908 - the average full-time CMO base cited earlier on this page - the federal statutory employer tax is $14,757, or 6.53% of salary: $11,439 of Social Security, $3,276 of Medicare, and $42 of FUTA. At $400,000 the same calculation gives 4.32%. The rate nearly halves between a director-level salary and a well-paid CMO, and the reason is one capped line item.

4. Putting the two halves together, honestly

Statutory taxes are the floor, not the load. The rest - paid leave, health insurance, retirement contributions, supplemental pay - is where most of the 43.0% lives, and that part is genuinely a benchmark rather than a quote. Three caveats that matter more than the headline:

• ECEC is an economy-wide average, and an executive package is not an average package. Executive compensation is weighted toward bonus and equity, which the ECEC wage and benefit lines treat differently from base salary. Use 43.0% to sanity-check a budget, not to price an offer.
• The statutory sub-component of that markup is regressive. Because of the Social Security cap, the payroll-tax share of the load shrinks at CMO salaries, so the true executive markup is below the economy-wide 43.0% on the tax lines and can be above it on the insurance and equity lines. The average conceals movement in both directions.
• Equity is not in any of these numbers. ECEC measures cash and benefit cost per hour worked. Dilution from an option grant is a real cost to existing shareholders and it does not appear in a payroll figure at all.

What this changes about the comparison on this page: the corrected private-industry markup of 43.0% raises the estimated true cost of a $225,908 CMO relative to the old 28-35% figure, which makes the fractional comparison look better. That is a reason to be more careful with it, not less. The number that should actually drive the decision is not the load percentage but whether the role needs a full-time person at all - a question the hiring guide and its contract-terms section handle in more detail, and one the load math cannot answer.

Finally, the other side of the ledger. A fractional CMO invoicing as an independent contractor carries no employer payroll tax at all - no Social Security match, no Medicare match, no FUTA. That is not a discount the operator is granting you; it is a consequence of worker classification, and it comes with the classification risk and none of the paid leave, benefits continuity, or full-time availability that the loaded salary buys. Treating the payroll-tax saving as pure margin is the mirror image of the denominator error at the top of this section.

Employer cost questions, answered

Is the BLS 30% benefits figure a markup on salary?

No. BLS reports benefits as a share of total compensation, and total compensation already includes the benefits. In 2026 Q1 private industry, benefits were 30.1% of total compensation but $14.01 against $32.60 of wages, which is a 43.0% markup on wages. Using 30.1% as a markup understates the employer load by about 30 percent.

What is the employer payroll tax on a $225,908 CMO salary in 2026?

$14,757 in federal statutory tax, or 6.53% of salary. That is $11,439 of Social Security (6.2% on the first $184,500 of wages), $3,276 of Medicare (1.45% with no cap), and $42 of FUTA. State unemployment and workers' compensation are extra and vary by state.

Does the employer pay the 0.9% Additional Medicare Tax?

No. IRS Publication 15 states directly that "There is no employer share of Additional Medicare Tax". The employer must begin withholding it once an employee's wages pass $200,000 in a calendar year, but the tax is imposed on the employee alone. It is a withholding obligation, not an employer cost.

Why does the employer tax rate fall as the CMO salary rises?

Because Social Security is capped and Medicare is not. In 2026 the employer stops paying the 6.2% Social Security tax after $184,500 of wages, while the 1.45% Medicare tax continues on every dollar. Total employer federal tax is therefore 6.53% at $225,908 and 4.32% at $400,000 - the same absolute Social Security amount spread over a larger salary.

What is the Social Security wage base for 2026?

$184,500. IRS Publication 15 for 2026 states that "the social security wage base limit is $184,500", with the rate at 6.2% each for employer and employee. Wages above that level are still subject to Medicare tax, which has no wage base limit.

Does a fractional CMO cost the company any payroll tax?

No, provided the engagement is a genuine independent-contractor relationship. A contractor invoice carries no employer Social Security, Medicare, or FUTA. That saving is a consequence of worker classification rather than a concession on rate, and it is offset by the absence of paid leave, benefits, and full-time availability. Misclassifying an employee as a contractor moves those taxes back onto the company along with penalties.

How Long the Full-Time Seat Actually Lasts: The Cost Input Nobody Prices

Short answer: a full-time CMO is quoted as an annual salary, and an annual salary quietly assumes the person stays. The federal turnover record says the seat is shorter than that. In professional and business services, the sector that contains most in-house marketing leadership, the Bureau of Labor Statistics Job Openings and Labor Turnover Survey puts total separations at 4.36 percent of employment per month for January to July 2026, which annualises to 52.3 percent and implies an average job tenure of about 23.0 months. On this page's own loaded figure of $322,993 a year, the full-time decision is therefore a $617,746 commitment that ends, on sector-average turnover, inside two years, and every dollar of search and onboarding amortises over 23.0 months rather than over a career. None of the eight tables above this one priced that.

1. What this data is, and what it is not

JOLTS is a monthly establishment survey. It reports job openings, hires, quits, layoffs and discharges, and other separations, as both levels and rates, by industry supersector. It is the only federal source that measures how often jobs turn over in something close to real time, and it is free and machine-readable through the BLS public data API.

It is not an executive series. JOLTS has no occupational dimension at all: the professional and business services figure covers every job in the sector, from a chief marketing officer to a temporary staffing placement, and staffing agencies sit inside that supersector, which pulls its turnover rate up. Executive tenure is generally reported as longer than an all-occupations average. So the tenure figure below should be read as the low end of a plausible range for a CMO seat and not as a prediction about one. There is no federal series for chief marketing officer tenure, and any page quoting one precisely is quoting a private survey. This page would rather publish a sourced sector floor with its caveat attached than an unsourced executive number without one.

2. Turnover and vacancy by sector, January to July 2026

Two derived columns carry most of the meaning. The vacancy proxy is job openings divided by hires, the months of open positions outstanding for every month of hiring, and it is the standard way to read vacancy duration out of JOLTS. Implied tenure is the reciprocal of the annualised separations rate, which is the average tenure a workforce settles at when hiring and separations are in balance.

Source: U.S. Bureau of Labor Statistics, Job Openings and Labor Turnover Survey, seasonally adjusted, series JTS<industry>000000000<element><L or R>, retrieved from the BLS public data API on 17 September 2026. Openings and hires are monthly averages of January to July 2026, in thousands. Vacancy proxy and implied tenure are arithmetic on the BLS series and are not BLS publications.

The spread is the finding. A marketing leader hired into financial activities sits in a sector where the average job lasts 48.6 months; the same hire into professional and business services sits at 23.0. That is a factor of about 2.1, and it changes the amortisation of every fixed cost in the hire. It also cuts the other way: financial activities runs the longest vacancy proxy at 2.33 months, so the seat that lasts longest also takes longest to fill.

3. The market loosened, which argues against the usual case for hiring fractionally

The standard argument for a fractional engagement is that senior marketing talent is scarce and slow to hire. That was true in 2022. The record says it is much less true now, and a cost page that left this out would be selling rather than informing.

Source: the same JOLTS series, calendar-year averages for 2019 and 2022 and the January to July average for 2026. Implied tenure is the reciprocal of the annualised total separations rate.

Job openings in professional and business services averaged 2,149,000 a month in 2022 and 1,248,000 in 2026 to date, a fall of 41.9 percent. Implied tenure in the same sector lengthened from 18.9 to 23.0 months, meaning people are leaving less often. Across the whole economy the BLS unemployed-per-job-opening ratio went from 0.53 in 2022 to 1.03 in 2026 to date: roughly two openings for every unemployed person, then roughly one each. Hiring a full-time CMO in 2026 is measurably easier than it was in 2022, so the honest version of the fractional case has to rest on whether the role needs a full-time person at all rather than on whether you could find one. The security and compliance case is the clearest example of a role that is genuinely part-time by nature rather than part-time by necessity, and the hiring guide works through the test.

4. What a 23.0-month average tenure does to the cost math

Everything below is arithmetic on figures already established on this page plus the tenure and vacancy figures from Table 9. Nothing here is a survey result.

Loaded cost follows Table 4 above: wages plus the private-industry benefits markup of 43.0 percent derived from BLS Employer Costs for Employee Compensation. Tenure and vacancy follow Table 9. Search cost is not sourced and is expressed per $10,000.

Two things in that table are easy to misread, so they are worth stating plainly. The vacancy months are not a payroll cost. While the seat is empty the company saves the salary; what it loses is the leadership, and no federal series prices that. The honest claim is that on sector-average turnover the seat is covered 94.8 percent of the time and the coverage gap arrives at exactly the moment a handover is happening. And the search cost is deliberately blank. Executive recruiter fees are quoted privately and there is no primary federal source for them, so this page gives the amortisation per $10,000 of spend and lets you supply the figure your own search actually costs. A page that invented a recruiter fee to make its own comparison look better would be doing the thing the employer-load section above spends a thousand words correcting.

5. Method, cross-checks, and limits

Source: U.S. Bureau of Labor Statistics, Job Openings and Labor Turnover Survey, seasonally adjusted series, retrieved from the BLS public data API on 17 September 2026. Series identifiers follow the JOLTS convention of industry code, then state, area and size class, then the data element and the level or rate flag. The four supersectors used are total nonfarm (000000), professional and business services (540099), information (510000) and financial activities (510099), and the elements are job openings (JO), hires (HI), total separations (TS), quits (QU), layoffs and discharges (LD) and other separations (OS). The window pulled is January 2019 to July 2026, 91 monthly observations per series and 49 series in total. Anyone can reproduce the pull without an API key.

Cross-check one, internal consistency. BLS publishes total separations and its three components as four independent series. If they are consistent, quits plus layoffs and discharges plus other separations should equal total separations in every month. Checked across all four sectors and every month in the window: 364 of 364 month-sector pairs reconcile, 100.0 percent, with levels rounded to thousands. Nothing downstream would be trustworthy if that had failed.

Cross-check two, the employment base. Levels and rates are published separately, so the employment denominator can be recovered two independent ways: from hires level divided by hires rate, and from separations level divided by separations rate. The two recoveries agree to within 1.5 percent at the median and 4.8 percent at worst, the worst cases being small sectors where levels rounded to the nearest thousand lose precision.

Cross-check three, the assumption behind the tenure figure. Reading tenure as the reciprocal of the separations rate is only valid when hiring and separations are close to balanced, and that is testable in the same data. Over January to July 2026 the hires rate and the separations rate differ by 2.0 percent in professional and business services and 1.3 percent for total nonfarm, so the assumption holds well in the two sectors the headline figures come from. It holds least well in information, where the gap is 11.4 percent because that sector is still separating faster than it hires, which makes the information tenure figure the softest number in Table 9. It is labelled as such here rather than quietly used.

Limits, stated rather than buried. JOLTS has no occupational detail, so none of these figures is specific to marketing or to executives. Rates are ratios to employment rather than to headcount at a single firm, so they describe a labour market and not a company. The vacancy proxy assumes openings and hires are in rough steady state and it overstates duration when hiring is falling, which it is. Seasonally adjusted series are model output and get revised. And 2026 here is seven months of data, not a year. Every figure above is a market-level benchmark for sanity-checking a budget, the same standing this page gives the ECEC benefits markup, and not a quote for one hire.

Tenure and vacancy questions, answered

How long does a full-time CMO actually stay in the job?

There is no federal series for CMO tenure specifically, so the closest sourced answer is the sector rate. BLS Job Openings and Labor Turnover Survey data for January to July 2026 puts total separations in professional and business services at 4.36 percent of employment per month, which annualises to 52.3 percent. In a steady state that implies an average job tenure of about 23.0 months. That is an all-occupations sector average and not an executive figure, and executive tenure is usually reported as longer. Treat it as the floor of a plausible range rather than a forecast for one hire.

What does the average tenure figure do to the cost of a full-time CMO?

It converts an annual salary into a total commitment. On this page's loaded figure of $322,993 a year for a $225,908 salary, 23.0 months of employment costs $617,746. The salary line is the price per year and the tenure figure is how many years you get. A comparison that shows an annual salary against an annual retainer is comparing two prices without comparing what each one buys.

How much does the search cost add per month once it is amortised?

Every $10,000 of search, relocation and onboarding cost spread over an average tenure of 23.0 months adds $436 a month, or $5,229 a year. This page does not publish a search-cost figure because there is no primary federal source for one, so the number is given per $10,000 and you can multiply it by whatever your recruiter actually quotes.

Is the market for marketing leadership still tight in 2026?

No, and that argues against the usual case for hiring fractionally. Job openings in professional and business services averaged 2,149,000 a month in 2022 and 1,248,000 in January to July 2026, a fall of 41.9 percent. Across the whole economy there were 0.53 unemployed people per job opening in 2022 and 1.03 in 2026 to date. Implied tenure in the sector lengthened from 18.9 to 23.0 months over the same period. The scarcity argument was true in 2022 and it is much weaker now.

Does the sector a company sits in change the answer?

Substantially. Implied average tenure in 2026 to date is 23.0 months in professional and business services, 31.8 months in information and 48.6 months in financial activities, a spread of about 2.1 times between the fastest and the slowest. Financial activities also runs the longest vacancy proxy at 2.33 months against 1.25 in professional and business services. A services company and an insurer face genuinely different versions of the same hiring decision.

What the Full-Time Alternative Costs Where You Are: 332 Metros, Federal Data

Short answer: the full-time comparison on most cost pages, including the one above, runs on a national salary, and in 297 of 332 US metros the national figure overstates what a local hire is paid. The federal wage survey publishes a median for marketing managers in 332 of the country's 393 metropolitan areas. The median metro pays $134,990, 19.1% below the national median of $166,790, and only 35 metros reach the national figure. Loaded at this page's 43.0% benefits markup, the median manager in the median metro costs about $16,084 a month. That is a smaller gap to a fractional retainer than a national number implies, and this page would rather publish the smaller gap than the flattering one.

1. What this data is, and what it is not

Occupational Employment and Wage Statistics (OEWS) is the Bureau of Labor Statistics survey of what employers pay, by occupation and by place. It is the only federal source that publishes pay for a named occupation in every metropolitan area, and it is free through the BLS public data API. All 393 metro series for occupation 11-2021 were requested; 332 carry a 2025 median, two are marked "estimate not released" (Hattiesburg, MS and Rapid City, SD), and 59 small metros have no estimate for this occupation at all.

It is not a CMO series. There is no federal occupation code for chief marketing officer. Marketing managers is the closest line and it covers every seniority, so a median marketing manager is a floor for the person a CMO search would hire, not an estimate of them. The value of the metro cut is relative: it shows how much the full-time price moves with location, which a single national number hides.

2. The distribution across 332 metros

The spread is 2.88 times, from $80,200 in Morgantown, WV to $231,370 in San Jose-Sunnyvale-Santa Clara, CA. The middle half of metros sits between $121,990 and $154,620. Every wage below is a published median; the last two columns are this page's own arithmetic.

Source: U.S. Bureau of Labor Statistics, Occupational Employment and Wage Statistics, May 2025 release, occupation 11-2021 Marketing Managers, all 393 metropolitan statistical areas, annual median wage, retrieved from the BLS public API on 22 September 2026. The loaded monthly cost is arithmetic: the median times 1.430 (the private-industry benefits markup in Table 4), divided by 12.

3. The extremes rest on very few people

Before trusting a metro median, read how many people it is computed from. Six of the fifteen highest metro medians come from fewer than 300 employed marketing managers: Cheyenne 30, Dover 50, Greeley 100, Billings 160, Topeka 190 and Fort Collins 250. All ten of the lowest come from 100 people or fewer. A median from a few dozen respondents can sit far from what a real search in that metro would pay, in either direction. The large markets are the stable ones: New York (54,730 employed, $192,840), Boston (11,940, $213,910) and San Francisco (11,170, $220,480).

Source: U.S. Bureau of Labor Statistics, Occupational Employment and Wage Statistics, May 2025 release, occupation 11-2021 Marketing Managers, all 393 metropolitan statistical areas, annual median wage, retrieved from the BLS public API on 22 September 2026. The loaded monthly cost is arithmetic: the median times 1.430 (the private-industry benefits markup in Table 4), divided by 12. Employment is the metro estimate for the same occupation and release.

4. Florida, metro by metro

MarkCMO is based in Cape Canaveral, inside the Palm Bay-Melbourne-Titusville metro, so the Florida cut is the local one. No Florida metro reaches the national median. Palm Bay-Melbourne-Titusville is the highest at $161,630 on 400 people; Miami-Fort Lauderdale-West Palm Beach, the largest pool at 7,530, pays $135,870, and seven smaller Florida metros pay more. Tampa's median is $159,990, but BLS did not release its employment figure.

Source: U.S. Bureau of Labor Statistics, Occupational Employment and Wage Statistics, May 2025 release, occupation 11-2021 Marketing Managers, all 393 metropolitan statistical areas, annual median wage, retrieved from the BLS public API on 22 September 2026. The loaded monthly cost is arithmetic: the median times 1.430 (the private-industry benefits markup in Table 4), divided by 12.

5. What the metro data does to the cost comparison

Loaded at 43.0%, a median marketing manager costs more than $8,000 a month in all 332 metros, more than $10,000 in 331, more than $15,000 in 228, and more than $20,000 in only 32. Set that against Table 1: an embedded retainer of $8,000 to $15,000 a month buys about 10 to 20 hours a week. Per month of cash the fractional seat is usually cheaper; per hour it is not, and it never was. Against a full CMO salary, the savings quoted at the top of this page hold. Against the local marketing manager a smaller company might hire instead, the gap is narrower, it exists only when the role does not need 40 hours a week, and your metro decides how narrow it is.

How these figures were checked. Every median is a multiple of $10, as BLS publishes them. For the 48 metros in Tables 13 and 14, the annual median equals the published hourly median times 2,080 within $12.80, inside the rounding of the two figures. And the 18 large metros that also appear on this site's hiring guide were pulled again, independently, and all 18 medians match the 19 September pull to the dollar.

Limits. OEWS pools three years of survey responses, so a release is not a one-year reading. Workers are counted where they are employed, not where they live. Wages exclude bonus and equity. The 43.0% markup is an economy-wide private-industry average applied for comparison, not a quote for one package, the same standing Table 4 gives it. Metro definitions follow the current BLS delineation of 393 areas.

Metro cost questions, answered

What does a full-time marketing manager cost in my metro area?

Usually less than the national figure suggests. Across the 332 US metros where the federal wage survey publishes a median for marketing managers, the median metro pays $134,990, against a national median of $166,790. The range runs from $80,200 in Morgantown, WV to $231,370 in San Jose-Sunnyvale-Santa Clara, CA. Loaded with this page's 43.0% benefits markup, that is $9,556 to $27,567 a month. The source is BLS Occupational Employment and Wage Statistics, May 2025, occupation 11-2021.

Is the national salary figure a fair benchmark for a local hire?

In most of the country, no. Only 35 of 332 metros pay a median at or above the national $166,790, because the national figure is pulled up by a few large, high-wage markets. In the other 297 metros a cost comparison built on the national median overstates the full-time option. Use your own metro's median, and check how many people it rests on: six of the fifteen highest metro medians come from fewer than 300 employed marketing managers.

Is a fractional CMO retainer cheaper than a full-time marketing manager?

Per month, often; per hour, no. Loaded at 43.0%, the median marketing manager costs more than $8,000 a month in every one of the 332 metros, more than $15,000 in 228, and more than $20,000 in 32. An embedded retainer of $8,000 to $15,000 buys roughly 10 to 20 hours a week, not 40, so the saving is real only when the role does not need a full-time person. A CMO also sits above the median manager, so these figures are a floor.

What do marketing managers earn in Florida metros?

Less than the national median in every one of them. Of Florida's 22 metros, 21 publish a figure. The highest is Palm Bay-Melbourne-Titusville at $161,630, the Brevard County metro that includes Cape Canaveral, and the lowest is Gainesville at $114,150. Miami-Fort Lauderdale-West Palm Beach employs the most, 7,530 people, at a median of $135,870, below Palm Bay, Tampa, Jacksonville, Destin, The Villages, Orlando and Naples.

What Companies Actually Spend on Marketing: 2,123 SEC Filings, Not a Survey

Short answer: across the 1,228 SEC registrants that disclosed advertising expense for calendar 2025, the median company spent 1.59% of revenue on advertising, and across the 948 that disclosed all-in selling and marketing expense the median was 18.01%. On the 449 companies that disclosed both, the all-in figure is 11.56 times the advertising figure, and it is the larger of the two in 442 of the 449. Media is the small part of what a company spends to sell itself; the rest is overwhelmingly payroll, including the sales organisation that the all-in tag bundles in. That is the case for taking the leadership decision at least as seriously as the media plan, and it is what a single blended marketing-percentage benchmark hides.

1. What this data is, and what it is not

Every figure in this section comes from company filings with the U.S. Securities and Exchange Commission, read through the SEC XBRL frames API on data.sec.gov. These are numbers a company reported in an audited financial statement and tagged in the us-gaap taxonomy, not answers to a marketing survey. Two tags do the work. AdvertisingExpense is media and advertising only. SellingAndMarketingExpense is the all-in line, salaries included. The point of this section is that those two are nowhere near each other.

It is not a private-company benchmark, and the smallest band is not a proxy for one. Every company here is an SEC registrant. Below about $50M of revenue that population is genuinely unusual: many are recently listed, many are foreign-domiciled, and marketing intensity inside that band runs from effectively zero to almost all of revenue. Those rows are published with their full spread in Table 18 for exactly that reason. The bands above $250M are ordinary operating companies and the readings there are the ones to lean on.

The all-in tag bundles sales with marketing, and that biases one number in a direction worth naming. SellingAndMarketingExpense includes the sales organisation, so it overstates marketing alone. Where this section divides a retainer by that figure, the denominator is too big and the resulting share is therefore a floor. A marketing-only denominator would make the retainer a larger share of the function, not a smaller one. Saying so is cheaper than being caught by it.

Disclosure is elective. Of the 4,372 registrants reporting revenue for calendar 2025, only 2,123 (48.6%) broke out any marketing or advertising line at all. A company that considers marketing immaterial does not have to publish it, so this is a self-selected sample of slightly under half the market. That is a limit on every filing-based marketing benchmark, including this one.

2. Media is the cheap part of marketing, by 11.56 times

The cleanest way to see it is to look only at the 449 companies that disclosed both tags, so the comparison is the same set of businesses measured two ways. Their median advertising spend is 1.93% of revenue. Their median all-in selling and marketing spend is 22.32%. The all-in figure is the larger one for 442 of the 449, so this is a structural fact about how marketing money is spent, not an artefact of comparing two different samples or of one lopsided median.

Source: SEC XBRL frames API, data.sec.gov, us-gaap tags AdvertisingExpense and SellingAndMarketingExpense, USD, calendar year 2025 frame, retrieved 26 September 2026. Percentiles are of the ratio of the tagged expense to the same filer's reported revenue. Revenue uses RevenueFromContractWithCustomerExcludingAssessedTax where present and the older Revenues tag otherwise.

3. Above $50M, media intensity barely moves with company size. The function does.

Advertising intensity is almost perfectly flat across every band above $50M of revenue: 1.38% to 1.56%, a spread of 0.18 percentage points across 1,006 filers running from $50M companies to the largest in the country. Above $50M, a media budget really does scale roughly in proportion to revenue.

That flatness stops below $50M, and section 5 shows it clearly, so the two readings are worth reconciling here rather than leaving them to look contradictory. Inside the band this page actually sells into, media intensity moves a great deal: median advertising spend falls from 3.59% of revenue at $1M to $5M to 1.14% at $30M to $50M. So the claim is bounded, not general. Media intensity is flat among established companies and falls steeply across the small end.

The all-in function does not behave like either. It peaks in the $250M to $1B band at 21.25% of revenue and then falls monotonically with scale, to 16.87% at $1B to $10B and 10.11% above $10B, so the $250M to $1B band carries 2.10 times the marketing burden of the largest companies. Operating leverage in marketing is real, but the filings put it late: it arrives after $1B of revenue, not at $50M.

Source: as Table 15. Bands are by the filer's own reported revenue for the same period. The under $50M band is reported for completeness and is the least representative of the five, for the reasons given in section 1.

4. In calendar 2025 the all-in line fell and the media line did not

Holding the set of companies fixed removes the risk that a move in the median is really a change in who filed. On the 905 companies that disclosed all-in selling and marketing expense in both calendar 2024 and calendar 2025, the median fell from 18.57% of revenue to 17.88%, and 528 of them fell against 377 that rose. On the 1,209 that disclosed advertising in both years, the median went 1.55% to 1.58% and the direction was close to a coin flip, 582 up against 612 down.

Read together, those two rows locate where the reduction happened rather than merely asserting one. The all-in line fell while the media line did not, so the money came out of the part of the all-in line that is not media, which is predominantly payroll and the cost of carrying a sales organisation. It is a modest move at the median, 0.69 percentage points, and it is the direction of 528 companies against 377, so it is a tilt rather than a collapse. That is the environment a fractional engagement is bought in, and it is the honest context for the tenure and vacancy data earlier on this page.

Source: as Table 15, with the calendar 2024 frame for the prior year. Matched means the same filer identifier appears with a valid ratio in both periods. The counts of rose, fell and unchanged are of individual filers, not of the median.

5. The retainer against the function it would direct

Table 2 near the top of this page prices a retainer by revenue band. This table puts the same bands next to what SEC filers of that size actually spend on their whole selling and marketing function, so the retainer can be read as a share of the budget it would direct rather than as a number on its own. At $1M to $5M the retainer is 13.6% to 21.7% of the function. By $30M to $50M it is 4.4% to 7.1%. The share falls as revenue rises, which is the same direction as the advertising column in the same data: median advertising intensity drops from 3.59% to 1.14% of revenue across these five bands.

The counter-test belongs in the same table, so here it is. In every band, some filers spend less on their entire selling and marketing function than the low end of the retainer this page recommends for their size: between 4.7% and 11.1% of filers, depending on the band. For roughly one company in ten at these sizes a fractional retainer is not a slice of the marketing budget, it is bigger than the whole thing. A cost page that published only the flattering ratio would be hiding the group this engagement is wrong for.

Source: the retainer columns are this page's own Table 2, unchanged. The spend columns are SEC XBRL SellingAndMarketingExpense and AdvertisingExpense for calendar 2025, filers with revenue inside each band, 193 filers on the advertising measure and 283 on the all-in measure. The share column is arithmetic: the annualised retainer divided by the band's median all-in spend. The spread column is the full range across filers in that band, published because the median alone would hide how wide it is.

6. Method, cross-checks, and limits

Two checks were run before any of the above was published, because a pipeline that reads the wrong XBRL cell produces numbers that look perfectly reasonable.

Check one: do recognizable companies land where an informed reader would expect? They do, and that is the whole value of the test. Procter and Gamble comes out at 10.92% of revenue on advertising and Coca-Cola at 11.26%, both consistent with decades of disclosure from two of the largest advertisers in the world. The marketplaces that live on paid acquisition sit far higher, Etsy at 27.43% and Expedia at 26.47%. Walmart, which sells on price and footprint rather than advertising, comes out at 0.76%. A parsing error would not reproduce that ordering.

Source: as Table 15. Company names are as filed with the SEC. These twelve are shown as a method check and are not a sample of anything.

Check two: is the small-company band just pre-commercial life science? That was the obvious failure mode, and it is not what is happening. Of the 283 filers with revenue between $1M and $50M on the all-in measure, 10 (3.5%) carry a life-science name. Removing them moves the median from 14.72% to 14.12%, a difference of 0.60 percentage points. The band's behaviour is not a biotech artefact. It is, as section 1 says, an odd population for other reasons.

Limits. A calendar-year XBRL frame groups each filer's annual figure into the calendar year it best matches, so fiscal years are not perfectly aligned: 876 of the 1,228 advertising filers have a period ending in December 2025 and the rest are spread across other month ends. Filers reporting a tag but no usable revenue figure are excluded (206 on the advertising measure, 86 on the all-in measure), as are filers whose reported ratio exceeds 100% of revenue (14 and 65) and those reporting zero or negative revenue (7 and 10). Every exclusion is a count, never an estimate: no value in this section is imputed, interpolated or filled. Medians and percentiles are used throughout rather than means, because a handful of filers report near-zero marketing spend and would drag an average. Tags are elective and their scope varies between filers even within one tag. Nothing here is a forecast.

Marketing budget questions, answered

What percentage of revenue should go to marketing?

There is no single number, and the filings show why. Across the 1,228 SEC registrants that disclosed advertising expense for calendar 2025, the median spent 1.59% of revenue on advertising, with the middle half between 0.50% and 4.85%. Across the 948 that disclosed all-in selling and marketing expense, which includes the people, the median was 18.01%, with the middle half between 7.52% and 32.28%. Which figure applies depends entirely on whether you mean media only or the whole function, and the two differ by more than ten times. Source: SEC XBRL frames API, us-gaap tags AdvertisingExpense and SellingAndMarketingExpense, calendar 2025, retrieved 26 September 2026.

Is a fractional CMO retainer a large share of a marketing budget?

At the sizes this page prices, it is the minority of it, and the share falls as revenue rises. Setting this page's own retainer bands against the median all-in selling and marketing spend of SEC filers in the same revenue band, the retainer runs 13.6% to 21.7% of the function at $1M to $5M of revenue, and 4.4% to 7.1% at $30M to $50M. The honest counterpoint sits in the same data: between 4.7% and 11.1% of filers in each band spend less in total on selling and marketing than the low end of the retainer for their size. For roughly one company in ten at these sizes a fractional retainer is not a share of the marketing budget, it is larger than all of it.

Did companies cut marketing spend in 2025?

The all-in line fell and the media line did not. On the 905 companies that disclosed all-in selling and marketing expense in both calendar 2024 and calendar 2025, the median fell from 18.57% of revenue to 17.88%, and 528 fell against 377 that rose. On the 1,209 companies that disclosed advertising expense in both years the median barely moved, 1.55% to 1.58%, with 582 rising and 612 falling, which is close to a coin flip. Because the media line held while the all-in line dropped, the reduction came out of the non-media part of that line, which is predominantly payroll and the cost of a sales organisation. At 0.69 percentage points on the median it is a tilt, not a collapse.

Do most public companies disclose what they spend on marketing?

No, and it matters for every benchmark built from filings. Of the 4,372 SEC registrants that reported revenue for calendar 2025, 2,123, or 48.6%, disclosed any marketing or advertising line item: 1,249 used AdvertisingExpense, 1,023 used SellingAndMarketingExpense, 336 used MarketingAndAdvertisingExpense and 128 used MarketingExpense. Breaking marketing out is elective when it is not material, so this analysis, like every other one drawn from filings, describes the self-selected half that chose to publish. Any single company can be checked directly at data.sec.gov through the XBRL company facts or frames API.

Why Growth-Stage Companies Hire a Fractional CMO

The demand for senior marketing leadership has never been higher -- and the cost of getting it wrong has never been steeper. Yet most growth-stage companies face the same impossible math: a full-time Chief Marketing Officer costs $280,000 to $450,000 in year one including salary, benefits, equity, and recruiting fees, but the company is not yet at the scale to justify it.

A Fractional CMO solves this precisely. You get the same strategic capability -- go-to-market strategy, ICP definition, brand positioning, demand generation architecture, pipeline systems, and team leadership -- at $8,000 to $20,000 per month. The $150,000 to $300,000 in annual savings goes directly into paid media, content, product, or your next hire. For companies between $500K and $20M in revenue, this is the highest-ROI marketing investment available.

What a Fractional CMO Delivers

This is not advisory. This is not a slide deck and a handshake. A fractional CMO engagement with MarkCMO means a working operator embedded in your business, owning your marketing function, managing your team and agency relationships, and accountable to the same pipeline and revenue KPIs a full-time CMO would own.

• Go-to-Market Strategy: Precise ICP definition, competitive positioning, messaging architecture, and channel selection -- built for your specific competitive landscape and buyer behavior
• Demand Generation Architecture: Multi-channel pipeline engine covering SEO, content marketing, paid media, email nurture, and outbound -- built as compounding systems, not one-off campaigns
• Team and Agency Leadership: C-suite management of your marketing team, agency partners, and freelancers with board-ready reporting on pipeline, CAC, and marketing ROI
• Sales and Marketing Alignment: Joint pipeline reviews, lead quality SLAs, and revenue attribution so every marketing dollar is tracked to closed-won revenue
• Marketing Operations: CRM configuration, attribution modeling, marketing tech stack optimization, and performance dashboards that replace gut feeling with data
• Recruiting and Talent Development: When the company is ready, Mark recruits and onboards the full-time marketing leader who takes over the function

Industries and Market Context

The US B2B market is anchored by SaaS, Healthcare, Manufacturing, and Professional Services. Each vertical carries its own marketing complexity -- regulatory constraints in healthcare, long enterprise sales cycles in B2B tech, intense price competition in logistics, and procurement-committee dynamics in manufacturing and defense. A fractional CMO who has operated across all of these verticals accelerates results by months compared to a generalist who needs a full year to understand your buyers.

With over 6 million employer businesses in the United States and intense competition across every vertical, companies that invest early in marketing strategy and execution compound their advantages. Those that defer fall further behind. The fractional CMO model is purpose-built for this window -- when you need a senior strategic operator but can not yet justify a $350,000 full-time hire.

Learn more about hiring a fractional CMO

• What a fractional CMO actually does
• Fractional CMO cost calculator
• Fractional CMO, defined
• A fractional CMO's first 90 days
• Fractional CMO vs a marketing agency
• The ROI of a fractional CMO

The 90-Day Quick Start: What Happens When You Engage

Every MarkCMO engagement follows a structured 90-day framework designed to deliver measurable results fast while building the marketing system that compounds for years. There is no six-month discovery phase. No ramp time. You see results in the first 30 days.

Every engagement includes weekly leadership check-ins, monthly board-ready reporting, and a marketing system designed to produce pipeline independently of ongoing fractional oversight -- because the goal is never dependency, it is transformation.

Case Study: B2B SaaS: ARR Growth Accelerated to 3x in 12 Months

*Case study is representative of outcomes. Client details anonymized per NDA. Results vary by company size, market, and execution quality.

See more outcomes: Results & Case Studies

What Clients Say

Read all client testimonials →

About Mark Gabrielli -- Fractional CMO

Learn more: About Mark  |  Results and Case Studies  |  Fractional CMO Services  |  How to Measure Fractional CMO ROI