What Is a Fractional CMO?

Fractional CMO Definition

The term fractional refers to the engagement model, not the level of commitment or seniority. A fractional CMO brings the same strategic depth and executive authority as a full-time CMO - applied at a fraction of the cost.

Unlike a marketing consultant who delivers recommendations, a fractional CMO executes. They sit in leadership meetings, manage the marketing team, own the budget, and are accountable to pipeline and revenue outcomes.

What Does a Fractional CMO Do?

• Develops and owns the go-to-market strategy
• Manages the marketing team, agencies, and vendors
• Sets KPIs and holds the team accountable to pipeline metrics
• Reports to the CEO and board on marketing performance
• Leads demand generation, brand, and product marketing
• Prepares the company for fundraising or exit
• Builds the marketing function from the ground up

Why the Fractional Model Exists: What the Tenure Data Shows

The fractional CMO model exists because the full-time seat turns over faster than the work it owns. Spencer Stuart's CMO Tenure study, published in January 2026 and covering 346 named chief marketing officers at S&P 500 companies as of 30 June 2025, puts average CMO tenure at 4.1 years against 5.0 years for all C-suite roles at the same companies. Close to a third of the index, 31%, no longer staffs an enterprise CMO at all. Those two numbers, not a cost argument, are the honest case for fractional marketing leadership.

The number is routinely used to prove something it does not prove. "CMOs last four years" is usually offered as evidence that the role is impossible, that boards are impatient, or that marketing cannot demonstrate value. The same study undercuts that reading with its own data: 62% of departing CMOs move into a similar or larger role, 9% become chief executives, another 13% take a divisional CEO, president or COO seat, and 77% are at a new company inside six months. That is a pattern of mobility, not a pattern of failure. If you are reading a short stint on a marketing leader's resume, the base rate says ask where they went before you assume why they left.

What the data does establish is narrower and more useful. The seat is the least stable senior seat in the building, and it is unstable for structural reasons that a better hire does not fix. A brand repositioning, a category entry, a channel rebuild or a pricing change is a two- to four-year arc. A tenure of 4.1 years means the person who starts that arc is, on the average case, unlikely to be the person who finishes it. The problem a company is actually solving is continuity of senior marketing judgment, and hiring one full-time executive is only one of the available answers to it. The 31% of the S&P 500 running without an enterprise CMO are answering it a different way.

One caveat that matters more than the headline. This is S&P 500 data. A 30-person B2B software company is not a smaller version of an index constituent, and the 4.1-year figure does not transfer to it. What transfers is the structural observation, and it lands harder at small scale: a large company that loses its CMO still has a director bench, a research function, an agency roster and a documented plan. A small company that loses its one senior marketer usually loses the strategy with them, because the strategy was never written down anywhere else. That is the case for treating documented, transferable marketing leadership as the deliverable, whether the person delivering it is full-time, fractional, or an interim hire. It is also why the full-time comparison below is a question about structure and risk, not only about monthly cost.

Source: Spencer Stuart, "CMO Tenure 2026: Snapshot of an Expanding Role for Marketing Leaders", published January 2026. Sample: 346 named chief marketing officers at S&P 500 companies, measured as of 30 June 2025. Figures verified against the published study on 2 September 2026. Where a figure on this page carries an "as of" date, that date says when the underlying data was sourced; the review date says when the page was last checked against it.

Where Marketing Leaders Actually Are: Federal Data by State (as of September 2026)

Half of the country's marketing managers work in five states, and nine states have fewer than 1,000 of them at any level. The federal wage survey counts 395,240 marketing managers in the United States. California, New York, Texas, Florida and Illinois employ 50.4% of them while holding roughly 37.6% of all jobs. At the other end, Wyoming reports 100, North Dakota 180 and South Dakota 220. That clustering is the geographic half of the case for fractional marketing leadership: the tenure data above says the seat turns over, and the state data says that in most of the country the local pool to refill it from is small. Source: BLS Occupational Employment and Wage Statistics, May 2025 release, pulled 21 September 2026.

Density tells the same story as headcount. New York employs marketing managers at 2.19 times the national rate per job and the District of Columbia at 1.95, while 14 of the 48 reporting areas sit below half the national rate. Pay follows the dense markets: the median runs from $98,750 in West Virginia to $212,020 in Massachusetts, and only 11 areas reach the national median of $166,790. A company in a thin, lower-wage market can hire cheaply only if a qualified local candidate exists, and the counts in Table 13 say that often one does not.

How these figures were checked. Three tests were run against the published series before any number was used. The annual mean equals the hourly mean times 2,080 in all 48 areas, within $14.80. A single constant fits the location quotient divided by employment per 1,000 jobs in all 48 areas within the rounding of the published figures, implying about 2.54 marketing managers per 1,000 jobs nationally, so the state density figures are internally consistent. And the mean exceeds the median in 47 of 48 areas, the right-skewed shape expected of a wage distribution; the exception is Wyoming, where the estimate rests on 100 people.

Limits of this data. OEWS pools three years of survey responses, so one release is not a one-year reading. There is no federal CMO occupation code: marketing managers is the closest line and covers every seniority, so these are counts of the whole pool, not of CMO-ready candidates. Workers are counted where they are employed, not where they live, so remote staff are attributed to their employer's establishment. BLS did not release employment for Montana or wages for Delaware, and publishes no estimate for this occupation in Rhode Island or Vermont; those states are left out rather than estimated. The all-jobs share is derived from each state's published employment per 1,000 jobs and covers the 48 areas that report it. Wages exclude benefits, equity and bonuses.

Marketing talent by state, answered

Where do most marketing managers in the United States work?

Half of them work in five states. The federal wage survey counts 395,240 marketing managers nationally, and California (54,660), New York (54,000), Texas (44,680), Florida (24,820) and Illinois (21,140) together employ 50.4% of them, against roughly 37.6% of all jobs in the reporting states. The ten largest states hold 66.4%. The figures are BLS Occupational Employment and Wage Statistics, May 2025 release, for occupation 11-2021.

How many marketing managers are there in a small state?

Very few. Nine states report fewer than 1,000 marketing managers of any seniority: Wyoming 100, North Dakota 180, South Dakota 220, Alaska 240, West Virginia 300, Hawaii 570, Mississippi 770, New Mexico 800 and Louisiana 940. Together they employ 4,120, fewer than Utah alone. Nineteen states report fewer than 2,000. Those are whole-occupation counts, so the people with CMO-level experience are a small fraction of each.

Does a marketing leader cost less in a lower-wage state?

The local median is lower, when a local candidate exists. The median marketing manager earns $98,750 in West Virginia and $212,020 in Massachusetts, a 2.15x spread, and only 11 of 48 areas sit at or above the national median of $166,790. The catch is supply: the lowest-wage states are also among the thinnest pools, so a company that cannot hire locally recruits from the dense markets and pays their rates.

Why does geography matter when deciding on a fractional CMO?

Because the full-time option depends on who lives nearby or will move. Senior marketing talent is concentrated: half the national pool is in five states and fourteen areas sit below half the national density. A fractional leader works across clients remotely, so the choice stops being limited to the local pool. That is an editorial reading of the federal data, not a claim the data makes on its own.

Fractional CMO vs. Full-Time CMO

A full-time CMO costs $280,000-$420,000 per year including salary, benefits, and equity. A fractional CMO costs $3,000-$15,000 per month, providing the same strategic leadership at 20-40% of the cost.

The right choice depends on company stage. Companies under $15M ARR rarely have enough work - or budget - to justify a full-time CMO. A fractional CMO delivers executive leadership while preserving capital.

Full comparison: Fractional CMO vs. Full-Time CMO →

Fractional CMO Engagement Models

A fractional CMO is not one fixed arrangement. The role is structured to fit the company, and the structure decides the cost, the cadence, and how much the CMO owns. As of July 2026, these are the four engagement models companies most often use.

Cost ranges by company stage are in the fractional CMO cost benchmark. Model choice is a framework, not a rule; the right structure depends on your stage and goals.

When Should You Hire a Fractional CMO?

The typical trigger points for hiring a fractional CMO:

• Revenue between $500K and $15M and growth has plateaued
• The CEO is running marketing and hitting their ceiling
• Preparing for Series A, B, or growth equity fundraise
• Existing marketing team lacks senior strategic leadership
• Agencies are running without executive direction or accountability

How Much Does a Fractional CMO Cost?

Fractional CMO retainers range from $2,500 to $15,000 per month depending on scope and hours. Most engagements run $4,000-$10,000/month. There are no benefits, equity dilution, or recruiter fees.

Fractional CMO cost benchmark, as of July 2026 (market ranges by company stage):

A full-time CMO typically costs $200,000 to $350,000 per year in base salary plus bonus, equity, and benefits. A fractional CMO delivers the same executive leadership at 40-70% less, with measurable impact usually inside 30 to 60 days.

See full pricing breakdown →

What Does a Fractional CMO Actually Own?

The confusion about fractional CMO scope comes from comparing the role to marketing consultants or agency partners. A fractional CMO is an embedded executive -- they own the function, not a project.

A fractional CMO owns the full marketing function: ICP definition, positioning, messaging architecture, channel strategy, team direction, budget allocation, vendor management, agency oversight, and revenue attribution. They report to the CEO and are accountable to the same metrics a full-time CMO would be: pipeline generated, CAC, MQL-to-SQL conversion, and revenue influenced by marketing.

The distinction matters because companies often confuse "getting marketing help" with "getting CMO leadership." A marketing consultant delivers a strategy deck. A marketing agency runs campaigns. A fractional CMO builds the architecture that determines whether either of those investments work.

Fractional CMO vs. Agency vs. Marketing Consultant

Three of the most common options for companies that need marketing leadership but aren't ready for a full-time C-suite hire:

What Industries Use Fractional CMOs?

Fractional CMOs are most common in B2B companies, growth-stage startups, and regulated industries where marketing complexity outpaces the company's ability to hire a full-time CMO. The most active industries include:

• B2B SaaS -- Long sales cycles, complex buying committees, and the LTV/CAC model require CMO-level thinking, not just campaign execution.
• Healthcare and MedTech -- HIPAA constraints, multi-stakeholder buying, and clinical credibility requirements need a CMO with domain knowledge.
• Fintech -- Regulatory content requirements, enterprise financial buyers, and compliance review processes require a CMO who understands both marketing velocity and compliance constraints.
• Professional Services -- Thought leadership, referral network development, and trust-based selling require strategic marketing leadership, not just campaign management.
• PE-Backed Companies -- Aggressive growth targets on compressed timelines require an experienced operator who can produce results faster than a full-time hire search allows.
• Pre-Series A and Series A Startups -- Need marketing credibility for fundraising without the full-time cost that consumes capital before product-market fit is fully proven.

Where Full-Time Marketing Leadership Exists, Industry by Industry: Federal Data (OEWS May 2025)

Short answer: full-time marketing managers are concentrated in a handful of industries and nearly absent from several of the largest industries by employment. Across 25 large US industries covering 86,097,330 jobs, 55.4% of all US wage and salary employment, the density of marketing managers runs from 0.11 per 1,000 jobs in food services and drinking places to 18.54 per 1,000 at corporate head offices, a more than 150-fold range, against a national rate of 2.54. In 9 of the 25 industries there is less than one marketing manager per 1,000 jobs. Those 9 industries hold 64.2% of the jobs measured here and 11.5% of the marketing managers. That gap, not a cost argument, is the practical reason fractional marketing leadership exists: in large parts of the economy there is no full-time marketing leader to replace.

The list above names the industries where fractional CMOs are commonly hired. This section measures the other side of the same question with federal data: how many full-time marketing managers each industry actually employs, what they are paid, and how many marketing staff each of them leads. It uses the Occupational Employment and Wage Statistics program, the federal survey of establishments, sampled over three years, that the earlier state section on this page also draws on, at the industry level rather than the state level.

Marketing managers per 1,000 jobs, 25 industries

Read the right-hand columns together. In food services one marketing manager exists for every 9,295 jobs, and the median one is paid $81,480, 49% of the national median of $166,790 for the occupation. In specialty trade contracting, the plumbers, electricians, roofers and HVAC firms that make up much of the small-business economy, the figure is one for every 2,834 jobs. At the other end, management consulting runs one for every 65 jobs and corporate head offices one for every 54. Across these 25 industries, pay for the role tends to rise with how common the role is: the correlation between the log of density and the median wage is 0.71.

How concentrated the role is

The concentration is the finding. The 9 industries below one marketing manager per 1,000 jobs carry 64.2% of the jobs in this table and 11.5% of its marketing managers. The two densest, head offices and management consulting, carry 5.5% of the jobs and 36.5% of the marketing managers. Head offices alone employ 52,430 marketing managers, 13.3% of every marketing manager in the country, which is the single most important thing to know before reading Table 20 as a picture of who has marketing leadership.

What the head-office number hides, and what it does not

OEWS counts jobs where they are physically located, by establishment. A restaurant chain, a retailer or a hospital system that runs its marketing from a separate corporate office has those marketing managers counted under head offices, not under restaurants, retail or hospitals. So the thinness of food services or general merchandise retail in Table 20 is partly an accounting artefact of where large companies seat their marketing teams, and this data cannot say how large that part is. What the artefact cannot explain is the single-location business. An independent HVAC contractor, a two-site dental practice or a regional trucking firm has no separate head office for a marketing manager to be counted in; if it employed one, that person would appear in its own industry row. For owner-run businesses in the thinnest rows of Table 20, the low figure is a fair description of reality: marketing leadership is usually the owner, an agency, or nobody.

Marketing staff per marketing manager

Nationally there are 2.28 marketing specialists for every marketing manager. In 13 of the 25 industries the ratio is above that, and the highest readings sit in industries that are also thin on managers: general merchandise retailers at 19.24, food services at 5.61, repair and maintenance at 5.21, specialty trade contractors at 4.30 and ambulatory health care at 4.04. Subject to the same head-office caveat, that is the shape of a team with practitioners and few strategic owners, a situation a fractional CMO is often brought in to fix. The sales column tells a different story about the same industries. Repair and maintenance carries 7.50 sales managers per marketing manager and truck transportation 4.40, against 1.61 nationally: in some of the thinnest industries, commercial leadership sits with sales rather than marketing.

What this means for a company deciding whether it needs a CMO

If your industry sits in the top rows of Table 20, a fractional CMO is usually a replacement for, or a bridge to, a full-time hire, because the full-time role is normal in your sector and candidates exist. If it sits in the bottom rows, a fractional CMO is more often the first senior marketing role the company has ever had, and the work looks different: building the function, choosing which channels to stop, and deciding what the first marketing hire should be, rather than running an existing team. Neither situation is evidence that a company should spend more on marketing. The table describes how industries are staffed today, and a staffing norm is not a performance benchmark.

Data and analysis by Mark Gabrielli, fractional CMO, 3 October 2026.

How to Evaluate a Fractional CMO

Not all fractional CMOs have the same background or capability. Before hiring, evaluate candidates across five dimensions:

• Operator experience, not advisor experience -- Have they been in the CMO seat before, with direct accountability for revenue outcomes? Advisors and consultants are not the same as operators.
• Industry relevance -- Do they understand your buying cycle, ICP, and competitive dynamics from direct experience, not general marketing knowledge?
• Revenue attribution capability -- Can they build or assess a revenue attribution model? If they can't connect marketing spend to closed revenue, they can't lead modern B2B marketing.
• Scope clarity -- Are they offering to own the function, or just provide strategy? A true fractional CMO should be willing to manage the team, own the budget, and be measured on pipeline.
• References with outcomes -- Can they provide client references who can speak to specific revenue outcomes, not just satisfaction with the relationship?

The First 90 Days of a Fractional CMO Engagement

A fractional CMO engagement structured correctly follows a consistent 90-day architecture regardless of industry or company stage:

Days 1-30 (Diagnostic): Full audit of current marketing performance, ICP definition, attribution model assessment, team capability review, channel analysis, and competitive positioning. No campaigns should launch before this phase completes.

Days 31-60 (Architecture): Build the strategic layer -- GTM architecture, channel mix selection, messaging framework, demand generation system design, CRM configuration, and revenue attribution model. Every execution decision in phase three flows from this foundation.

Days 61-90 (Activation): Launch the first campaigns based on the new architecture. Initial pipeline metrics begin flowing. Team is trained on the new playbook. Attribution dashboards are live. First board-ready marketing report is delivered.

Companies that skip the diagnostic phase -- going straight to campaign execution -- consistently generate lower-quality pipeline at higher CAC because the ICP and channel selection haven't been validated.

Full guide: The First 90 Days of a Fractional CMO →

Is a Fractional CMO an Employee or an Independent Contractor?

A fractional CMO is engaged as an independent contractor and paid on a Form 1099, not through payroll. That label is not self-executing. Two different federal tests decide the question, they are applied by two different agencies for two different purposes, and neither one is bound by what your contract calls the relationship. As of September 2026 the wage and hour test is the six-factor economic reality test at 29 CFR part 795, and the tax test is the IRS common law control test. A properly run fractional engagement passes both comfortably. The exposure is not the model. It is drift.

This matters at the moment of hiring, not later, because the facts that decide it are the ones you set up on day one: how many clients the CMO carries, who supplies the tools, whether there is an end date, and whether anyone is told what hours to work. Nobody renegotiates those after a notice arrives.

The Six Factors, Applied to a Real Fractional Engagement

The Department of Labor's rule lists six factors and is explicit that they are weighed together. The regulation states that no one factor or subset of factors is necessarily dispositive and that these six factors are not exhaustive. That matters here, because a fractional CMO reliably loses one of the six and still lands clearly on the contractor side of the other five.

The fifth factor is the one worth being honest about. Marketing leadership at a company that sells through marketing is close to the core of the business, so the integral-part factor tends to point toward employee no matter how the engagement is written. That is not a defect in the arrangement. It is what a totality test is designed to absorb, and it is also the reason a checklist that treats any single factor as decisive gives the wrong answer.

The IRS Test Is a Different Test, and It Is the Expensive One

The Department of Labor decides whether someone is owed overtime. The IRS decides whether you should have been withholding tax. They are separate questions with separate tests, and an engagement can be comfortable under one and awkward under the other. The IRS groups the evidence into three categories.

The consequence is set out plainly in the IRS guidance. A business that classifies an employee as an independent contractor with no reasonable basis for doing so may be held liable for employment taxes for that worker, under Internal Revenue Code section 3509. Relief provisions may apply where there was a reasonable basis and all required federal information returns were filed consistently, but the IRS states that relief does not determine a worker to be an independent contractor. It relieves the tax bill without settling the status. If either side wants the status settled, either the business or the worker can file Form SS-8 and ask the IRS to determine it, and the IRS cautions that it may take at least six months to receive a determination.

The Rule Is Mid-Change, and the Direction Favours the Fractional Model

The analysis in force is the 2024 final rule, published at 89 FR 1638 and effective 11 March 2024, which is the six-factor totality test in Table 6. On 27 February 2026 the Department of Labor published a proposed rule at 91 FR 9932 to rescind that analysis and replace it with the one the Department adopted on 7 January 2021, and to apply the same analysis to the Family and Medical Leave Act and the Migrant and Seasonal Agricultural Worker Protection Act as well. Comments closed on 28 April 2026.

The 2021 analysis is not simply an older version of the same thing. It designates two of the factors, the nature and degree of control over the work and the individual's opportunity for profit or loss, as core factors that, in the Department's words, typically carry greater weight in the analysis than any other factor. Where both point the same way, the Department's position is that there is a substantial likelihood that is the individual's accurate classification.

Read against Table 6, that would help a genuine fractional CMO rather than hurt them. The two factors that would carry the weight are the two a real fractional engagement wins most clearly, and the factor a fractional CMO reliably loses, whether the work is integral to the business, would stop being able to drag the answer. As of September 2026 none of this has taken effect. No final rule has followed the February proposal, so the six-factor test is what governs today and is what you should structure against.

The Honest Version: This Is Usually Not a Close Call

Anyone selling classification anxiety to justify a hiring decision is overselling it. A fractional CMO with several concurrent clients, their own entity, their own tools, a fixed monthly fee and no set hours is the paradigm independent contractor under either test, and the analysis is not difficult. Most companies asking this question already have the right answer and want confirmation.

Two things sharpen it. First, the wage and hour exposure is smaller than it sounds. The Fair Labor Standards Act mainly buys minimum wage and overtime, and a marketing executive would in most cases sit inside the executive or administrative exemption, so even an adverse finding often produces no overtime owed. Those thresholds moved this year: federal courts vacated the Department's 2024 exemption rule, and on 15 May 2026 the Department removed that text from the Code of Federal Regulations and republished the prior text at 91 FR 27833, restoring the salary level to $684 per week, or $35,568 per year, with the highly compensated employee threshold at $107,432 per year. The tax side carries no such exemption, which is why the IRS test is the one to structure against.

Second, the risk is created by drift rather than by the model. The engagements that get reclassified are the ones that quietly became something else: one client instead of several, forty hours instead of eight, direct reports, a seat on the org chart, no end date. If a fractional engagement is heading there, the right response is to convert it deliberately to a full-time hire, not to keep the invoice format and hope. The paperwork is not the thing being tested. The working relationship is.

Classification questions, answered

Is a fractional CMO an employee or an independent contractor? +

A fractional CMO is engaged as an independent contractor and paid on a Form 1099, not through payroll. That label is not self-executing. Two separate federal tests decide it: the Department of Labor's six-factor economic reality test at 29 CFR part 795, which governs minimum wage and overtime, and the IRS common law control test, which governs employment taxes and withholding. Neither is bound by what the contract calls the relationship. A properly structured fractional engagement, with several concurrent clients and no set hours, passes both.

Can a company be penalised for misclassifying a fractional CMO? +

Yes, and the tax side is usually the expensive one. The IRS states that a business which classifies an employee as an independent contractor with no reasonable basis for doing so may be held liable for employment taxes for that worker, with the amounts set by Internal Revenue Code section 3509. Relief may be available where there is a reasonable basis and all required information returns were filed consistently, but the IRS is explicit that relief does not determine a worker to be an independent contractor. It only relieves the tax liability.

Does a fractional CMO get benefits, PTO or a company email address? +

A fractional CMO should not be on the benefits plan and should not accrue paid time off. The IRS lists employee type benefits such as a pension plan, insurance or vacation pay as evidence of an employment relationship under the type of relationship category. A company email address is a smaller signal and is often granted for practical access reasons, but benefits are the one to refuse. The paperwork saying contractor does not survive facts that say employee.

Is the independent contractor rule changing in 2026? +

It is proposed to change and has not changed yet. The analysis in force is the 2024 final rule at 89 FR 1638, effective 11 March 2024, which weighs six factors as a totality of the circumstances. On 27 February 2026 the Department of Labor published a proposed rule at 91 FR 9932 to rescind that analysis and replace it with the one it adopted on 7 January 2021, which treats control and opportunity for profit or loss as core factors carrying greater weight, and to extend the same analysis to the Family and Medical Leave Act and the Migrant and Seasonal Agricultural Worker Protection Act. Comments closed on 28 April 2026. As of September 2026 no final rule has followed, so the six-factor test still governs.

Would a misclassified fractional CMO actually be owed overtime? +

Often not, which is why the wage and hour exposure is usually smaller than the tax exposure. The Fair Labor Standards Act mainly buys minimum wage and overtime, and a marketing executive would in most cases fall within the executive or administrative exemption. Those thresholds moved recently: federal courts vacated the Department's 2024 exemption rule, and on 15 May 2026 the Department removed that text from the Code of Federal Regulations and republished the prior text at 91 FR 27833, restoring the salary level to $684 per week, or $35,568 per year, with the highly compensated employee threshold at $107,432 per year.

What makes a fractional CMO engagement risky to classify? +

Drift, not the model itself. An engagement that starts at one day a week and ends up at forty hours for a single client, with direct reports, set hours, company-supplied tools, no other clients and no end date, is an employee wearing a different word, and every factor in both tests will read it that way. If either side wants certainty, either the business or the worker can file Form SS-8 and ask the IRS to determine the status formally, though the IRS warns it may take at least six months to receive a determination.

Who Owns the Work a Fractional CMO Creates?

By default, the fractional CMO does, not the company that paid for it. Copyright "vests initially in the author or authors of the work" under 17 U.S.C. 201(a), and the exception that moves authorship to the hiring party, a work made for hire, is far narrower than most marketing contracts assume. It reaches employees, and a well-structured fractional engagement is deliberately not employment. Past employees it reaches only nine enumerated categories of commissioned work, and the most valuable things a fractional CMO produces sit in none of them. What actually moves ownership is a separate, signed, written assignment under 17 U.S.C. 204(a). A company can hold every file, have paid every invoice, and have a work-for-hire clause in its contract, and still not own the copyright.

This is a summary of what the Copyright Act says, not legal advice, and it does not account for your contract, your state, or your facts. The point of putting it on this page is narrower than that: it is a question a company should ask before it engages a fractional CMO, almost nobody asks it, and the answer is not the one the contract template implies.

The section above closes the easiest route to ownership, and that is not a contradiction

The classification section on this page argues that a properly run fractional engagement is clearly an independent contractor arrangement, and that structuring it that way is how you stay out of a wage or employment tax problem. That conclusion is correct, and it is also exactly what removes the employee route to a work made for hire. Section 101(1) reaches "a work prepared by an employee within the scope of his or her employment". The more cleanly an engagement passes the six-factor economic reality test and the IRS common law control test, the more certainly it fails the test for the simplest form of client ownership.

These are the same fact with opposite consequences, and the right response is not to muddy the engagement. It is to notice that the two questions are settled by two different instruments: the working relationship settles classification, and a signed assignment clause settles ownership. A company that tries to fix the second by loosening the first has traded a solvable paperwork problem for an unsolvable payroll one.

The nine categories, and why the strategy work is never in them

Section 101(2) does not let any commissioned work be a work made for hire. It lists nine categories and stops: a contribution to a collective work, part of a motion picture or other audiovisual work, a translation, a supplementary work, a compilation, an instructional text, a test, answer material for a test, and an atlas. Copyright Office Circular 30 states the test as four conditions that must all hold: the work falls in one of the nine categories, there is a written agreement, it expressly says the work is a work made for hire, and all parties sign it. Failing any one of the four defeats it.

Read the list against what a fractional CMO is hired to produce and a pattern falls out that is the opposite of what most people expect. The production assets often do qualify: a video ad is plainly part of an audiovisual work, a benchmark dataset is a compilation, a training curriculum can be an instructional text. The strategic work never does. A positioning framework, a go-to-market architecture, a board narrative, the messaging that the whole company then repeats, none of these is a translation, a compilation, an atlas or any of the rest. The highest-value deliverable in the engagement is the one the work-for-hire clause cannot reach.

The practical reading of Table 9 is not that work-for-hire clauses are useless. It is that a clause which relies on work for hire alone is load-bearing exactly where it is weakest. A competent version says the work is a work made for hire where the law allows it, and that to the extent it is not, the author assigns all right, title and interest to the client. The second half is what does the work on the strategy documents, and it costs nothing to include.

Two things the file transfer does not do

First, handing over the file is not handing over the copyright. Section 202 says it in one sentence: "Ownership of a copyright, or of any of the exclusive rights under a copyright, is distinct from ownership of any material object in which the work is embodied." Getting the Figma file, the working document, the raw footage and the source files is an access question. It settles whether you can use and edit the material in practice. It settles nothing about who owns it.

Second, and pulling the other way, nobody owns the strategy itself. Section 102(a) protects "original works of authorship fixed in any tangible medium of expression", and 102(b) is categorical about the limit: "In no case does copyright protection for an original work of authorship extend to any idea, procedure, process, system, method of operation, concept, principle, or discovery." The positioning document is protected. The positioning is not. A company that wants to stop a departing fractional CMO from using the same approach at the next client is asking for something copyright does not sell, and should be looking at confidentiality and non-solicitation terms instead, which are a different negotiation with different limits.

The registration gap, and the $20 that is the wrong thing to optimise

Ownership and enforceability are separate. Under 17 U.S.C. 411(a) no civil infringement action can be brought until the claim is registered, and under 412 there is no statutory damages or attorney's fees award for infringement that started before registration took effect, unless registration is made within three months after first publication. Without those, a plaintiff is left proving actual damages and profits on a marketing asset, which is expensive and often not worth doing.

The figures make the arithmetic obvious. Statutory damages run from $750 to $30,000 per work, up to $150,000 where willfulness is proved, and down to $200 for an innocent infringer. Electronic registration is $45 for a single author, same claimant, one work, not for hire, or $65 on the standard application. There is a detail in that fee line worth pausing on: the $45 tier is expressly conditioned on the work not being a work made for hire. The cheapest registration path is the one where the fractional CMO is named as the author and has assigned the rights, which is also the path that survives Table 9. The $20 difference is not the thing to optimise, but it is a small confirmation that the assignment route is the normal one and the law is built to expect it.

The one asymmetry that is real and almost never mentioned

A work made for hire and an assignment are not equivalent even when both are available. Section 203 gives an author a right to terminate a grant during a five-year window opening 35 years after execution, or 40 years from execution where the grant covers publication, whichever ends earlier, and it does so "notwithstanding any agreement to the contrary". It applies "in the case of any work other than a work made for hire", and Circular 30 repeats the carve-out: these termination provisions do not apply to works made for hire. So the route that is usually unavailable is also the only one that is permanent.

The honest weight to put on that is small. A five-year window that opens in 2061 on a campaign asset written in 2026 is not a business risk any company should be pricing today, and most marketing work will have no commercial value by then. It is worth knowing for two narrow cases: a brand asset with a long life, such as a name, a mark's supporting artwork or a foundational piece of brand writing, and any situation where a buyer's counsel is running diligence on what the company actually owns. Pretending the asymmetry does not exist is what makes an otherwise careful page unreliable on the one question a lawyer will check.

What this means when you are the one hiring

None of this is a reason to hesitate over a fractional engagement. It is a reason to read four lines of the contract before signing it. Does it assign copyright, and not only recite work for hire? Does the assignment cover work product created during the engagement rather than only named deliverables? Is it signed by the individual who actually creates the work, and not only by an LLC that the individual happens to own? And is there a separate, explicit term handing over accounts, source files and platform access, which copyright does not touch at all? A fractional CMO who cannot answer those cleanly is telling you something about how many times they have done this.

For what it is worth, this is the part of an engagement where the fractional model is usually cleaner than the alternative. An agency relationship frequently ends with the agency retaining the underlying assets and licensing them back. A senior independent operator has no reason to hold anything, and assigning it all is the normal term. Ask anyway, and get it in writing, because under 204(a) writing is not a formality. It is the thing that makes the transfer valid.

Ownership questions, answered

Who owns the marketing assets a fractional CMO creates? +

By default the fractional CMO does. Under 17 U.S.C. 201(a) copyright vests initially in the author, and the work-made-for-hire exception reaches employees plus nine enumerated categories of commissioned work. A fractional CMO is an independent contractor, and positioning documents, strategy decks and ad copy fall in none of the nine categories. Ownership moves to the client through a signed written assignment under 17 U.S.C. 204(a), and through nothing else.

Does a work made for hire clause in a fractional CMO contract actually work? +

Partly, and only for some deliverables. Copyright Office Circular 30 sets four conditions that must all hold: the work falls within one of the nine statutory categories, there is a written agreement, it expressly says work made for hire, and all parties sign it. Video ads, compilations such as benchmark datasets, and some training material can qualify. Positioning frameworks, go-to-market strategy and most ad and web copy cannot. A clause that relies on work for hire alone leaves the most valuable work uncovered, which is why a well-drafted clause adds an assignment fallback.

If I have all the files, do I own the copyright? +

No. 17 U.S.C. 202 states that ownership of a copyright is distinct from ownership of any material object in which the work is embodied. Receiving the Figma files, the source documents and the raw footage settles whether you can practically use and edit the material. It transfers no copyright on its own. Those are two separate contract terms and a good engagement letter contains both.

Can a company own the marketing strategy itself? +

No, and no contract changes that. 17 U.S.C. 102(b) says copyright protection does not extend to any idea, procedure, process, system, method of operation, concept, principle or discovery. The document that sets out a go-to-market strategy is protected. The strategy is not. A company that wants to limit what a departing fractional CMO can reuse is asking a confidentiality and non-solicitation question, not a copyright one.

Do I need to register a copyright to enforce it? +

You need it to sue and to get the remedies worth suing for. Under 17 U.S.C. 411(a) no civil infringement action can be brought until the claim is registered, and under 412 statutory damages and attorney's fees are unavailable for infringement that began before registration took effect, unless registration is made within three months after first publication. Statutory damages run from $750 to $30,000 per work and up to $150,000 for willful infringement. Electronic registration is $45 for a single author, same claimant, one work, not for hire, or $65 on the standard application.

Can a copyright assignment be undone years later? +

In principle, yes, and the contract cannot prevent it. 17 U.S.C. 203 lets an author terminate a grant during a five-year window opening 35 years after execution, or 40 years from execution where the grant covers publication, whichever ends earlier, and it applies notwithstanding any agreement to the contrary. It does not apply to works made for hire. For most marketing assets this is a theoretical point, because the work will have no commercial value in 2061. It matters for long-lived brand assets and it is the kind of thing a buyer's counsel checks during diligence.

What a Marketing Budget Is Actually Made Of: Advertising Against Everything Else

Across the 508 US public companies that separately disclosed both figures for calendar 2024, the median company spent 9.5% of its selling and marketing budget on advertising. The other 90.5% went to everything else: sales and marketing staff, tooling, content, events and operations. That single split is the most useful thing a founder can know before pricing marketing leadership, because a fractional CMO is bought out of the second number, not the first. The retainer ranges in Table 2 above are priced into the non-advertising part of the budget, and this section measures how large that part usually is.

One caution belongs before the tables rather than after them, because it limits every figure here. The SellingAndMarketingExpense tag bundles the sales organisation together with marketing: salaries, commissions and sales operations sit inside it. So the 90.5% remainder is a ceiling on non-media marketing cost, never a measure of it, and it is certainly not "90.5% of marketing is people". A company with a large enterprise sales force will show a small advertising share for reasons that have nothing to do with how it markets. Read the number as the share of go-to-market spend that is bought media, which is exactly what it is.

What was measured, and one cut that the data refused

The population is every registrant whose calendar-2024 filing carries both an advertising figure and a selling-and-marketing figure for the same period: 1,210 filers reported selling and marketing, 1,631 reported advertising, and 566 reported both. Of those, 58 were dropped on stated exclusions, leaving 508. The exclusions are counts, not estimates: 38 had no usable revenue tag, 16 reported a non-positive advertising figure, and 4 reported advertising larger than selling and marketing, which means the two tags are not nested for that filer and no share can be computed from them.

A second cut was attempted first and then abandoned on the evidence, which is worth recording because it is the more obvious question. The intent was to split the reported SG&A line into its selling-and-marketing and general-and-administrative halves, so a reader could see how much of "overhead" is really go-to-market. That is not answerable from these filings. Only 83 of the 1,972 filers reporting SG&A also report selling and marketing separately, 4.2% of them, because the two are alternative presentation choices rather than a subtotal and its component. Worse, 8 of those 83 report selling and marketing larger than the SG&A line they supposedly sit inside, which proves the two tags do not share a scope for a measurable share of filers. A usable population of 75 companies out of 1,972, with a known scope contradiction inside them, is not worth publishing a median from. The advertising split below survives that test; the SG&A split does not.

The distribution matters more than its midpoint. From the 10th percentile to the 90th the advertising share runs 1.2% to 61.1%, a 50.9x spread, and the two ends do not represent better and worse companies. They are different business models. Of the 508 filers measured, 357, or 70.3%, spend under a quarter of go-to-market on media, and only 77, 15.2%, spend half or more. If you have been told that marketing budget means advertising budget, the filings say that is true for about one company in seven.

The spread inside an industry is far wider than the spread between industries

The industry a company is in explains much less about its marketing mix than the way it sells does. Inside business services, the largest group at 226 filers, the advertising share runs 1.5% at the 10th percentile to 59.9% at the 90th, a 39.9x spread. Between the six industry medians in Table 16 the spread is only 2.6x, from 5.6% to 14.8%.

That comparison has an obvious weakness and it was tested rather than left alone. The 2.6x between-industry figure is flattered by the suppression rule: requiring 12 filers per group leaves only six groups of the 45 in the data, and the groups it removes are the small extreme ones. Relaxing the threshold widens the between-industry spread considerably, as Table 17 shows, and the honest version of the claim uses the widest of those rather than the narrowest. Even when every industry group down to two filers is included, the between-industry spread reaches 17.4x, still well below the 39.9x found inside business services alone. The conclusion survives its own weakest assumption, which is the only reason it is stated here.

Scale changes the mix twice, and in opposite directions

Two things happen at the top of the size range and they pull against each other. The advertising share of go-to-market is flat across four revenue bands, 10.2%, 8.6%, 9.0% and 9.1% from under $50M through $5B, and then roughly doubles to 20.8% above $5B. At the same time selling and marketing as a share of revenue is flat through those same four bands, within 5.6 percentage points of each other, and then falls 2.1x, from 22.7% to 10.6%. The largest companies spend a much smaller share of revenue on go-to-market, and a much larger share of that smaller budget on bought media.

The doubling at the top could easily be three consumer names rather than a property of scale, so it was checked. Excluding every consumer-facing industry group from both bands, the top band still reports 18.2% on 37 filers against 8.9% on 101 filers at $1B to $5B. And it is not a handful of outliers pulling a median: 29 of the 41 filers above $5B, 70.7%, sit above the $1B-to-$5B median of 9.1%. The shift is broad.

The second half of that finding is worth flagging because this routine has now reached it three times from three different denominators. Marketing efficiency, measured as spend against revenue, does not improve gradually as a company grows through the stages a founder actually lives through. It is flat from $50M to $5B and then arrives in one step. A company planning a path from $80M to $800M should not budget on the assumption that its marketing share of revenue will fall along the way.

Two companies in the same four-digit industry, five times apart

The clearest single illustration in Table 19 is not the widest gap, it is the narrowest classification. Salesforce, Inc. reports 7.5% of go-to-market on advertising and Intuit Inc. reports 39.4%, 5.2x apart, and both file under the same four-digit SIC code, 7372, prepackaged software. Microsoft Corporation files at 7.0% in the same code, 5.7x from Intuit. None of the three is a badly run company. Salesforce sells six-figure contracts through an enterprise sales organisation, and Intuit sells to small businesses and consumers who have to be reached before they will buy. The classification is identical and the correct marketing budget is not comparable. Across the whole table the advertising share runs 1.5% to 81.1%, a 54.1x spread, among companies that are all large, all public and all competently run.

The mix did not move between 2024 and 2025

Matching the 447 filers that reported both figures in both calendar years, the median advertising share of go-to-market went from 9.5% to 9.3%. In total 212 companies raised the share and 228 lowered it, with 7 unchanged. That is a coin flip and it is reported as one. The full calendar-2025 population of 456 filers reads 9.4%, within a tenth of a point of the matched figure. This is the sort of year that gets narrated as a shift back to brand spending or a retreat from it, and on these filings it is neither.

What this does and does not tell you about hiring marketing leadership

For software specifically, which is where most of this page's readers sit, the picture is sharper than the all-industry median. Among the 124 filers classified as prepackaged software, the median company puts 6.9% of go-to-market into advertising and 93.1% into everything else, and 105 of the 124, 84.7%, spend under a quarter on media. Software also carries a heavier selling-and-marketing line than the business services group it sits inside in Table 16, 32.4% of revenue against 28.7%, though among the groups in that table it is second to instruments and medical devices at 37.3%. A large budget and a small media line is the normal shape of a software go-to-market, not an anomaly to be corrected.

Two limits of this evidence should be stated plainly, because both cut against the way a figure like this is usually deployed. First, AdvertisingExpense is a disclosure election, not a required line. Companies that fold advertising into selling and marketing without breaking it out are absent from this population entirely, and a company that separately discloses advertising is plausibly one that spends more of it. That biases the 9.5% median upward, so the typical media share across all companies is more likely lower than this figure than higher. Second, and repeating the caution at the top because it is the one most likely to be dropped in a quote: the remainder includes the sales organisation, so it cannot be read as a marketing headcount budget.

And the honest conclusion for the decision this page is about. None of these numbers tell you what your marketing budget should be, and a table of industry medians is close to the weakest evidence you could use to set one. That applies to this table as much as to anyone else's. What the filings do establish is narrower and more useful: the dominant line in most companies' go-to-market spend is not media, the variation between companies in the same industry is many times the variation between industries, and the thing that predicts the mix is how a company sells rather than what it sells. A defensible marketing plan is argued from your own motion, your own sales model and your own margins. If the strongest case anyone can make for a budget is that the sector median is a particular number, the case is weak, and that is as true when the number comes from us as when it comes from a vendor.