In-house vs agency cost, compared with your own numbers
In-house costs less when work fills a seat all year. Agencies cost less when volume swings. Load salary by about 1.46 and compare per-project costs.
An in-house hire costs less than an agency when the work fills the seat every week of the year. An agency costs less when volume swings or a skill is needed only in bursts. Salary is only part of the bill: in June 2026, benefits were 31.5 percent of employer costs for full-time private industry workers, a load factor of about 1.46 on wages. Run the comparison with your own numbers.
In-house versus agency cost
Change any input; the sentence updates as you type. The figures below the form are the worked example.
How it is worked out
in-house = (managers × manager salary + specialists × specialist salary) × (1 + benefits) + people × tools
agency = retainer × 12 + oversight × manager salary × (1 + benefits)
break-even retainer = (in-house − oversight cost) ÷ 12- benefits
- employer benefit costs as a share of salary: $17.03 of benefits per $36.97 of wages for full-time private-industry workers in June 2026, which is 46.1%
- tools
- software seats, equipment and training you pay for each person you employ
- oversight
- the share of a manager's year you spend briefing, reviewing and steering the agency
Worked example, with the default inputs:
- Salaries: 1 × $166,790 + 3 × $78,760 = $403,070.
- With benefits at 46.1%: $403,070 × 1.461 = $588,885. Tools: 4 × $10,000 = $40,000. In-house total: $628,885.
- Agency: $40,000 × 12 = $480,000, plus oversight 0.25 × $166,790 × 1.461 = $60,920. Agency total: $540,920.
- Break-even retainer: ($628,885 − $60,920) ÷ 12 = $47,330 a month.
| Input | Default | Source |
|---|---|---|
| Manager roles you would hire | 1 | Example input, not a benchmark. Replace it with your own. |
| Manager salary | $166,790 | U.S. Bureau of Labor Statistics27 August 2026 |
| Specialist roles you would hire | 3 | Example input, not a benchmark. Replace it with your own. |
| Specialist salary | $78,760 | U.S. Bureau of Labor Statistics27 August 2026 |
| Benefits on top of salary | 46.1% | U.S. Bureau of Labor Statistics9 September 2026 |
| Tools and overhead per person | $10,000 | Example input, not a benchmark. Replace it with your own. |
| Agency retainer | $40,000 | Example input, not a benchmark. Replace it with your own. |
| Your time to run the agency | 0.25 | Example input, not a benchmark. Replace it with your own. |
Embed this calculator on your site
<iframe src="https://www.cmotribune.com/embed/in-house-vs-agency" title="In-house versus agency cost, from CMO Tribune" width="100%" height="1200" style="border:0;max-width:760px" loading="lazy"></iframe>Is it cheaper to hire in-house or use an agency?
The short answer: it depends on how much work you have and how steady it is. In our reading, a full-time hire costs less when the work fills the seat every week of the year. An agency costs less when volume swings or a skill is needed only in bursts.
Start with the in-house side. The U.S. Bureau of Labor Statistics (BLS) put the mean annual wage for marketing managers at $177,770 in May 2025, with a median hourly wage of $80.19. Salary is only part of the bill.
U.S. Bureau of Labor Statistics, Table 1. National employment and wage data from the Occupational Employment and Wage Statistics survey by occupation, May 2025
For private industry workers in June 2026, wages and salaries accounted for 70.0 percent of employer costs. Benefits made up the remaining 30.0 percent. For full-time private industry workers, benefits were 31.5 percent.
U.S. Bureau of Labor Statistics, Employer Costs for Employee Compensation, June 2026, 9 September 2026
Then look at what marketers actually did. In a 2023 study by the Association of National Advertisers (ANA), 82 percent of ANA members reported having an in-house agency. Yet 92 percent of respondents also worked with an external agency.
Those respondents did an average of 61 percent of their work in-house. They sent work outside mainly for bandwidth or missing capabilities. We read that as most large marketers running both models at once rather than picking one.
A worked example
These are round example inputs, not benchmarks. Take a hire at a $150,000 salary. If wages are 70.0 percent of employer cost, $150,000 divided by 0.70 gives a loaded cost of about $214,286 a year.
Now compare an agency retainer of $180,000 a year for similar scope. The hire costs $214,286 and the retainer costs $180,000, a gap of $34,286 in the agency's favor.
The math changes with volume. If the agency bills $15,000 per project and you need 30 projects a year, the agency cost is $450,000. One salaried person who can absorb that load costs the same $214,286 either way.
| Scenario (example inputs) | In-house loaded cost | Agency cost |
|---|---|---|
| Steady scope on retainer | $214,286 | $180,000 |
| 30 projects at $15,000 each | $214,286 | $450,000 |
Run this with your own numbers before the budget meeting. Use your real salary offer, your benefits load and the agency's quoted scope of work. Then count how many months of the year the work would keep a person fully busy. Our marketing headcount benchmarks help you check whether the team size you are planning is in line with peers.
What is the fully loaded cost of a marketing employee?
Fully loaded cost is everything the company spends to keep one marketer working for a year, divided by nothing. Start with the salary and add each line below. The order matters less than the rule: count every cost once.
Fully loaded cost = base wage × benefits load + tools + recruiting (spread over expected tenure) + management time + vacancy cost
| Input | What it covers | Where the number comes from |
|---|---|---|
| Base wage | Salary plus any bonus you expect to pay | Your offer letter or pay band |
| Benefits load | Health insurance, retirement, paid leave, legally required benefits | Bureau of Labor Statistics multiplier, or your payroll provider |
| Payroll taxes | Employer share of taxes on wages | Sits inside the legally required benefits line, so do not add it twice |
| Tools | Seats and licenses in your martech stack, laptop, software | Your finance team's per-seat costs |
| Recruiting | Agency fees, job ads, interview hours, onboarding | Last three hires, averaged |
| Management time | Hours a manager spends directing and reviewing | Manager's loaded hourly cost × hours per week |
| Vacancy | Lost output while the seat is empty | Months to fill × monthly value of the role |
The benefits load
The Bureau of Labor Statistics (BLS) reported that total employer compensation costs for full-time private industry workers averaged $54.00 per hour worked in June 2026, against wages of $36.97, which gives a load factor of about 1.46. Benefits made up the rest, at $17.03 per hour. Two pieces of that are worth knowing. Legally required benefits were $3.76 per hour, and paid leave was $4.40 per hour. Because paid leave is already in the multiplier, do not add vacation weeks again elsewhere.
U.S. Bureau of Labor Statistics, Employer Costs for Employee Compensation, June 2026, 9 September 2026
TrinityP3, a marketing management consultancy, puts the add-on for benefits and taxes at 30 to 45 percent onto base salary. It also puts onboarding at up to 30 percent of first-year salary.
TrinityP3 Global Marketing Management Consultants, In-House vs Marketing Agency: Costs, Pros/Cons & Decision Guide, 5 January 2026
A worked example
These are example inputs, not benchmarks.
- On a base of $100,000, the TrinityP3 range of 30 to 45 percent puts loaded pay at $130,000 to $145,000.
- Using the BLS factor, $100,000 × 1.46 = $146,000.
- Tools at $15,000 per year.
- Recruiting at 30 percent of a $100,000 salary is $30,000; spread over a 4-year tenure, that is $7,500 per year.
Then add the last two lines from your own records. We think management time and vacancy are the ones finance teams most often leave out, and they are the two you can measure only inside your company. Compare the result with your team shape in marketing headcount benchmarks.
The hidden costs on both sides of the comparison
Both columns of the comparison carry costs that never appear on a salary line or a rate card. Our side-by-side below pairs them so you can price each one with your own numbers.
McKinsey's survey of 40 chief marketing officers and senior vice presidents, fielded June 25 to July 1, 2021, found that agile ways of working place outsized strain on agency relationships. Only 3 percent called the transition "smooth". The practices that helped included scopes of work and fee arrangements. In our reading, those two areas are where agency hidden costs collect.
McKinsey & Company, When agile marketing breaks the agency model, 29 September 2021
| Hidden cost | In-house team | Agency |
|---|---|---|
| Getting people in place | Recruiting fees and a recruitment gap and knowledge transfer when someone leaves, plus vacancy time | Agency absorbs hiring, but not everyone on the account is fully dedicated |
| Tools and add-ons | Software seats, with licensing, integration, maintenance and training | Outside purchases billed at a standard 20-25% markup |
| Changing the work | Reassigning internal staff | Ad-hoc changes require scope requests; on fixed fees, changes go through a change order billed separately |
| Management time | Non-billable time for training and internal meetings, plus your managers' hours | Account management layers, and client time on formal briefing documents and contractual reviews |
Scope is the line to watch on the agency side. When priorities shift, McKinsey found that the uncertainty makes it difficult for partners to predict the resources they need. We think that uncertainty tends to surface as change orders.
McKinsey & Company, When agile marketing breaks the agency model, 29 September 2021
A worked example
These are round example inputs, not benchmarks.
- On a $180,000 annual retainer, $15,000 of change orders adds about 8.3 percent to the fee.
- $100,000 of outside purchases at a 20 percent markup adds $20,000.
- On the in-house side, price vacancy as the months a role sits open multiplied by the monthly cost of covering it.
Treat older rate studies as a starting point. Before you sign, write change-order rules into the scope of work and check them against our agency contract checklist.
How many companies run in-house agencies, and what they keep outside
The Association of National Advertisers (ANA) has surveyed its members on in-house agencies every five years since 2008. The latest round was fielded in February and March of 2023, with 162 respondents.
The share of ANA members with an in-house agency rose from 42 percent in 2008 to 58 percent in 2013 and 78 percent in 2018.
ANA members with an in-house agency, 2008 to 2018
| Label | Value |
|---|---|
| 2008 | 42% |
| 2013 | 58% |
| 2018 | 78% |
Adoption rose at every survey. The 2023 share cited earlier was up from 78 percent in 2018, 58 percent in 2013, and 42 percent in 2008. The ANA predicts penetration will eventually peak at 85 to 90 percent.
Workload keeps rising
In 2023, 88 percent of respondents indicating their in-house shops’ workload increased in the past year, and 67 percent said it rose "a lot." In 2018 the figure was 90 percent. The top challenges were managing workflow (increased projects), scaling efficiently/managing resources, and project prioritization.
In our reading, a cost model built on last year's volume will understate next year's in-house headcount. Price the team at the volume you expect, not the volume you have.
What moved inside, and what stayed out
Over three years, 65 percent of respondents have moved some established business from external agencies to their in-house team.
| Area | What the ANA found | What to test in your model (our reading) |
|---|---|---|
| Digital creative | Most moved: social media, search and email | Volume is steady enough to staff for |
| Traditional creative | Moved: print collateral, direct mail, internal communications, out-of-home and radio | Check whether production tools add to the in-house cost |
| Media | 54 percent of in-house agencies handle some media planning or buying | Holdouts called media too complex; keep specialist fees in the agency column |
How in-house teams are judged
Cost savings remained the top key performance indicator (KPI), but it decreased in importance from 2018 to 2023 (69 to 62 percent). Business performance increased significantly in importance from 2018 to 2023 (45 to 59 percent).
We think a board case built only on rate savings now covers half of what peers measure. Add a second line to the scorecard that ties in-house output to revenue or pipeline, using incrementality tests where you can, and agree that line with finance first. Our guide on CMO, CEO and CFO alignment sets out how.
A worked example and the break-even point
Here is the method with round numbers. They are an example, not a benchmark. Replace each one with your own before the figure goes to your CEO or CFO.
Step 1: salaries. Take a three-person team: one lead at 150,000 and two specialists at 100,000 each, which is 200,000 for the pair. Base pay for the team is 150,000 plus 200,000, or 350,000. Check the lead's figure against the U.S. Bureau of Labor Statistics (BLS) median hourly wage for marketing managers, $80.19 in May 2025, and against your own pay bands.
U.S. Bureau of Labor Statistics, Table 1. National employment and wage data from the Occupational Employment and Wage Statistics survey by occupation, May 2025
Step 2: load. In June 2026, BLS put total compensation for full-time private industry workers at $54.00 per hour worked against wages and salaries of $36.97. That is a ratio of about 1.46. Applied to the team, 350,000 times 1.46 gives 511,000.
U.S. Bureau of Labor Statistics, Employer Costs for Employee Compensation, June 2026, 9 September 2026
Step 3: the costs payroll misses. Add 15,000 a year for tools. Set recruiting fees and vacancy cover at 15,000 each, which comes to 30,000. The in-house total is 511,000 plus 15,000 plus 30,000, or 556,000.
Step 4: the agency side. Assume an annual retainer of 180,000 that covers 30 projects under its scope of work. That works out to 180,000 divided by 30, or 6,000 per project.
Step 5: break-even. Divide the in-house total by the agency's cost per project: 556,000 divided by 6,000 is about 92.7. At roughly 93 projects a year, the two options cost the same. Below that volume the agency costs less. Above it the team does, as long as three people can produce that volume.
| Line | Example input | Example result |
|---|---|---|
| Base salaries | 150,000 + 2 × 100,000 | 350,000 |
| Loaded pay | 350,000 × 1.46 | 511,000 |
| Tools, recruiting, vacancy | 15,000 + 30,000 | 45,000 |
| In-house total | 511,000 + 45,000 | 556,000 |
| Agency cost per project | 180,000 ÷ 30 | 6,000 |
| Break-even volume | 556,000 ÷ 6,000 | about 93 projects |
The last condition carries the weight. We think the break-even number is a capacity question as much as a cost one. If your three people can deliver 60 projects a year, they never reach 93, and the agency stays cheaper at every volume they can handle.
What the calculator asks for and returns
The calculator takes six inputs:
- Salary for each role
- A load factor, which defaults to the BLS ratio
- Annual tool spend
- Recruiting and vacancy cover
- The agency retainer
- The number of projects the retainer covers
It returns four outputs:
- The loaded in-house total
- The agency's cost per project
- The break-even volume
- The in-house cost per project at your current volume
Run it twice. First use last year's actual project count. Then use the count in next year's plan. If the break-even point falls between the two, that gap is what you take to the board.
Where each input comes from and how to read the result
Each field in the calculator has one owner. Pull every number from that owner and write the source next to it. The CFO conversation then becomes a review of inputs, not a debate about the answer.
| Field | Where it comes from | What to check (our reading) |
|---|---|---|
| Benefits and payroll costs | BLS employer cost data, the series cited earlier | Use the full-time private industry figures, not all workers |
| Tools and software seats | Your finance team's license records | Include seats the agency supplies today that you would have to buy |
| Recruiting | Finance or talent acquisition, last 12 months of search fees and job ads | Average across several hires, not the most recent one |
| Agency fees | Signed contracts, the retainer and scope of work | Add billed markups and change orders, not just the base fee |
Salaries
BLS gives national medians by occupation. The median annual wage for marketing managers was $166,790 in May 2025. For market research analysts, the median annual wage was $78,760 in May 2025. Treat these as a floor for the model. Your offer letters, if you have recent ones, are the better input.
Bureau of Labor Statistics, Advertising, Promotions, and Marketing Managers, 27 August 2026
Bureau of Labor Statistics, Market Research Analysts, 27 August 2026
How to read the result
Break-even is the point where the annual in-house cost for a defined scope equals the annual agency spend for that same scope. As an example, not a benchmark: if the loaded in-house cost is $180,000 and the agency spend is $150,000, the in-house option runs $30,000 above break-even.
- Above break-even: the agency is cheaper on cost alone. An in-house case then has to rest on speed, control or knowledge kept inside the company, and you should say so plainly.
- Below break-even: in-house is cheaper on cost. Check whether the team has the capacity to absorb the scope without new hires a year later.
- Close to break-even: the answer depends on your weakest input. Rerun it with the recruiting and tool figures your finance team signs off on.
Context for the CFO
Gartner's 2026 CMO Spend Survey covered 401 CMOs and other marketing leaders. It found that "labor is claiming a larger share of marketing budgets". Labor's share of the total marketing budget rose from 21.9% in 2025 to 24.5% in 2026.
We think this changes how a hire lands with finance. A CFO who already sees people costs rising will want the in-house case framed as a shift of spend from agency fees to salaries, not as new spend. Most respondents reported annual revenue of over $1 billion, so a smaller company should treat the shift as direction rather than a target. For more on the survey, see our digest of Gartner's 2026 spend findings. For the framing, see the guide to CMO, CEO and CFO alignment.
Common mistakes and the steps to take next
Most comparisons that fail with finance fail on inputs, not arithmetic.
| Mistake | What it does to the result | Fix (our reading) |
|---|---|---|
| Salary-only costing | Makes in-house look cheaper than it is | Use the loaded cost from the earlier section |
| Ignoring vacancy months | Assumes every seat is filled all year | Price the gap at agency or freelance rates |
| Comparing unlike tiers | Sets strategy fees against production rates | Sort work into tiers before pricing |
| Leaving out management time | Understates the agency side | Add client-side briefing and review hours |
Here are the first and last errors as an example, not a benchmark. Take a salary of 100,000 against a loaded cost of 150,000, and the salary-only view hides 50,000. On the agency side, add 15,000 of client-side management time to a fee of 180,000 and the true agency cost is 195,000. Both corrections move the answer, in opposite directions.
The tier error needs the most care. In Cella's model, the outside firm takes strategic work, such as image advertising and competitive analysis, labeled "Tier 1 (and above) work." The in-house team produces Tier 2, "execution against established creative," and Tier 3, "production design." An agency rate for Tier 1 thinking and an in-house rate for Tier 3 production are not the same product.
cellainc.com, In-House Versus Agency Rates, 19 September 2011
Steps to take next
- Sort the work into tiers. List last year's projects and tag each by tier and by medium, which is the split Cella recommends before deciding what to own.
- Build functional rates. If your team does not charge back, Cella suggests calculating rates by function, because that is easiest to compare to external providers.
- Get bids on real projects. Ask two or three firms to bid on two or three tightly scoped projects, such as a simple ad, a campaign and a brochure or website. Our agency RFP template and scope of work definition help keep the bids comparable.
- Run the numbers by tier. Compare loaded in-house rates with agency rates inside each tier, then add vacancy and management time.
- Decide what stays outside. In the ANA study, 92% of respondents still worked with an external agency. Their reasons were extra bandwidth when the internal team was too busy and work the in-house shop lacked the capabilities for. We think those two reasons make a sound test for each Tier 1 line.
- Revisit the rates. Cella notes it may be necessary to recalculate rates once or twice a year as labor and overhead change.
cellainc.com, In-House Versus Agency Rates, 19 September 2011
Marketing Dive, ANA: In-house agency trend continues to gain steam, 8 May 2023
What this page does not cover
This page does not cover how to choose a specific agency, pay data outside the United States, or the employment and tax rules that apply to hiring. It uses example inputs, not benchmarks, and it is not financial or legal advice.
Related on CMO Tribune: Marketing team structure by revenue band: functional, pods, hub-and-spoke and business-unit models; Marketing team size benchmark, as a share of headcount and by revenue band; Marketing mix modeling, explained for the CMO who has to choose a measurement model.
Questions
Is it cheaper to have an in-house marketing team or an agency?
It depends on how much work you have and how steady it is. In an example with round inputs, a $150,000 hire loads to about $214,286 a year, so a $180,000 retainer for similar scope is $34,286 cheaper. If you need 30 projects a year at $15,000 each, the agency costs $450,000 and the hire is cheaper.
What is the fully loaded cost of a marketing employee?
Fully loaded cost is base wage times a benefits load, plus tools, recruiting spread over expected tenure, management time and vacancy cost. In June 2026, the Bureau of Labor Statistics (BLS) put total employer cost for full-time private industry workers at $54.00 per hour worked against wages of $36.97, a load factor of about 1.46. Payroll taxes and paid leave already sit inside that multiplier, so do not add them twice.
How much does a marketing manager cost a company per year?
The BLS put the mean annual wage for marketing managers at $177,770 in May 2025 and the median at $166,790, with a median hourly wage of $80.19. Salary is only part of the cost. Apply a load factor of about 1.46, then add tools, recruiting, management time and vacancy. Your own offer letters and pay bands are the better input.
What percentage of companies have an in-house agency?
In a 2023 Association of National Advertisers (ANA) study, 82 percent of members reported having an in-house agency, up from 78 percent in 2018, 58 percent in 2013 and 42 percent in 2008. The ANA predicts penetration will peak at 85 to 90 percent. Most still use outside help: 92 percent of respondents also worked with an external agency.
How do I calculate break-even between in-house and agency marketing?
Divide the loaded in-house total by the agency's cost per project. In a round example, a three-person team costs 556,000 a year and a 180,000 retainer covering 30 projects works out to 6,000 per project, so break-even is about 93 projects. Below that volume the agency costs less. Check that your team has the capacity to deliver the break-even volume at all.
What marketing work should stay with an outside agency?
ANA respondents sent work outside mainly for bandwidth when the internal team was too busy and for capabilities the in-house shop lacked. In Cella's model, the outside firm takes strategic Tier 1 work such as image advertising and competitive analysis, while the in-house team handles execution and production design. Holdouts on media called it too complex, though 54 percent of in-house agencies handle some media planning or buying.
Sources
- Table 1. National employment and wage data from the Occupational Employment and Wage Statistics survey by occupation, May 2025, U.S. Bureau of Labor Statistics
- Employer Costs for Employee Compensation, June 2026, U.S. Bureau of Labor Statistics, 9 September 2026
- In-House Agencies No Longer A Trend: They're Here To Stay (ANA Report), ANA, 2 May 2023
- In-House vs Marketing Agency: Costs, Pros/Cons & Decision Guide, TrinityP3 Global Marketing Management Consultants, 5 January 2026
- When agile marketing breaks the agency model, McKinsey & Company, 29 September 2021
- In-House Versus Agency Rates, cellainc.com, 19 September 2011
- The Continued Rise of the In-House Agency: 2023 Edition, ana.net
- Advertising, Promotions, and Marketing Managers, Bureau of Labor Statistics, 27 August 2026
- Market Research Analysts, Bureau of Labor Statistics, 27 August 2026
- Gartner Marketing Survey Finds Awareness and Conversion Account for 62.6% of Total Media Spend, Gartner, 8 June 2026
- ANA: In-house agency trend continues to gain steam, Marketing Dive, 8 May 2023
Written by the CMO Tribune Desk and edited by Lila Odinaev. Every factual sentence was checked against a passage on the source it cites. See how we work, or report an error on the corrections page.