Hire an agency when the plan is already clear and you need more work shipped than your people can ship. Hire a fractional CMO when competent vendors and staff are producing output that nobody is steering, and all the steering currently routes through you. Hire in-house when the work is continuous, specific to your brand, and you can carry four to six months of hiring plus ninety days of ramp before it pays anything back.
Those three sentences settle most of these decisions. Founders still get it wrong because they shop for a solution before naming the bottleneck, so they buy throughput for an ownership problem, then conclude the category failed.
I’ve hired agencies, managed them, consolidated four into one, and fired ones doing perfectly good work against the wrong objective. At Ancestral Supplements the move from $28M to $60M+ came out of a leadership rebuild rather than a channel change. Dr. Gabrielle Lyon’s audience passed a million and the practice passed $6M with a content operation built in-house on purpose, because that work was proprietary. Paul Saladino MD’s content-to-commerce engine needed production volume, a different problem with a different answer.
The three-bottleneck test
Before you price anything, spend a week noticing where your marketing gets stuck. There are three places it stalls, each with a different fix.
Strategy bottleneck: activity without a shared definition of success
Your calendar is full of marketing. Ads are running, emails are going out, somebody launched a TikTok push last month. Ask three people what winning looks like this quarter and you get three answers, none of which is a number. Your agency reports ROAS, your email vendor reports open rates, and neither number tells you whether contribution margin moved.
The symptom you’ll recognize in your own week: you keep approving things that are individually reasonable and you can’t say what any of them add up to.
Throughput bottleneck: the plan is clear and nothing ships fast enough
You know the offer, the audience, and the next three campaigns. What you don’t have is enough hands. Creative takes eleven days to turn around, the landing page for the September launch is still a doc, and your best ad set has been running the same four assets since June.
Symptom: your meetings are about status rather than direction. Everybody’s behind and nobody’s confused.
Ownership bottleneck: competent vendors, no connective tissue
Paid media agency, email agency, a freelance designer, maybe a coordinator. Each does good work inside their box. No one holds the whole number, so every handoff and every prioritization call comes to you. You’re the integration layer, and you’re also the CEO.
Symptom: you can name what each vendor is doing this month and you cannot name who is accountable if the quarter misses. When two agencies disagree about attribution, you’re the tiebreaker, which is a job you didn’t hire for.
Most stalled $3M to $15M brands have an ownership bottleneck wearing a strategy bottleneck’s clothes. The plan is missing because nobody senior enough is being paid to make one and defend it.
Fractional CMO vs agency vs in-house vs consultant
| What it fixes | Who owns the number | Ramp time | Monthly cost | What breaks | |
|---|---|---|---|---|---|
| Fractional CMO | Ownership, then strategy | The fractional CMO | 2-4 weeks | $8K-15K | Fails without real decision rights |
| Agency | Throughput and channel craft | You, or whoever sits above them | 30-60 days to launch, 90 to judge | $8K-20K per channel | Optimizes its own channel with nobody above it |
| In-house hire | Continuous, proprietary work | Depends on seniority | 4-6 months to hire, 90 days to ramp | $12K-30K loaded | Mid-level hire with no leader above them becomes a coordinator |
| Consultant | A specific answer to a specific question | You | Days to weeks | Project fee, or $5K-10K retainer | Recommendation sits in a doc, nobody owns execution |
Costs are what I see across health and wellness DTC in 2026. The agency line is per channel, which founders forget when they compare a $12,000 fractional retainer to a $9,000 agency retainer. Two agencies and a freelancer runs $22,000 a month with nobody above them. The fractional bands are broken out further in what a fractional CMO costs.
Where each option honestly wins
Agencies win on throughput and channel craft, and it isn’t close. A good paid media team has run more accounts this quarter than your in-house hire will run in a career. They have creative pipelines, platform reps, benchmark data across dozens of brands, and can double output in two weeks without you posting a job. When I’ve consolidated vendors I’ve kept agencies more often than I’ve replaced them. What gets blamed on agencies is usually a management failure above them.
In-house wins when the work is continuous, proprietary, and load-bearing. Brand voice, customer insight, retention lifecycle. Anything where institutional memory is the asset and a 90-day vendor handoff would destroy it. Lyon’s content operation belonged in-house because the voice was the product. If a function will still exist in three years and touches your customer relationship directly, you want it on payroll, and you should be willing to pay for the ramp.
A fractional CMO wins when decisions queue behind you. Someone with executive standing renegotiates scopes, cuts a channel, sets one definition of success, and holds vendors to outcomes instead of deliverables. That work has a ceiling. Above roughly $15M, one or two days a week isn’t enough presence and you want the seat full time.
A consultant wins when you have a bounded question. Should we move off Amazon-first? Are our subscription economics broken, or is acquisition broken? You want an answer, and you have the team to execute it. More on that scope in what an ecommerce consultant actually does.
The hybrid that works most often at $3M to $15M
The structure I build for stalled brands in this band has three layers, each option used for what it’s good at.
Fractional operator on top. Owns contribution margin or a revenue target. Sets the quarterly plan, the budget allocation, and the measurement standard. Runs the vendor relationships with hire-and-fire authority over the marketing function.
Agencies underneath for throughput. Paid media, often creative production. They report to the fractional CMO, not to you, against scopes the operator wrote.
One or two in-house owners for what can’t be outsourced. Usually brand and customer insight in one seat, retention lifecycle in another. They report to the fractional CMO on a dotted line and to you formally, which matters when the engagement ends and they stay on.
Reporting lines are the whole thing. If the agencies still report to you while a fractional CMO advises from the side, you’ve bought a consultant at operator prices and kept the bottleneck.
How each pairing fails
Agency reporting to a founder with no time. The agency does what it can measure, which is its own channel. Nobody reallocates. Blended acquisition cost drifts up over three quarters while every channel report looks fine. This is the most common failure I walk into.
In-house hire with no leadership above them. You hire a marketing manager at $95K because a CMO is out of reach. Six months later they’re coordinating the same vendors you were, escalating the same decisions, and quietly looking for a job where somebody can teach them something. Good person, wrong structure.
Fractional CMO with no decision rights. Every call still routes back to you for approval, so the retainer buys a very expensive second opinion. Founders who land here usually aren’t being controlling on purpose. They’ve never had someone they trusted to make the call, so the handoff never happens. If you can’t name three decisions you’d let this person make without you, wait.
Consultant with no execution capacity. The deck is right and nothing moves, because the team already at capacity now has a strategy document on top of it.
Price the outcome, not the month
Comparing monthly fees is how brands end up with the cheapest structure that can’t produce the result. Run it the other way.
Cost per outcome: total monthly spend on the structure, divided by the specific change you’re buying.
A brand at $6M spending $22,000 a month on two agencies and a freelancer with nobody above them is paying $264,000 a year for execution against a plan no one owns. Add a $12,000 fractional retainer and the annual number goes to $408,000, which sounds worse until you price what the top layer does: kills one channel, redirects that budget, sets a measurement standard, hires the retention owner. If that sequence moves contribution margin four points on $6M, it paid for itself twice over.
The same math kills a lot of engagements, which is why I run it early. Under about $2M there’s rarely enough margin for a leadership layer, and the honest answer is an agency plus your own attention. Above $15M you’re paying fractional rates for a seat that needs to be full.
Switching without going dark for a quarter
Changing structure is where brands lose momentum, usually because they cut before they build. Sequence it.
- Bring the new layer in while the old one runs. Overlap for 30 to 45 days. Paying two vendors for six weeks costs less than a dead quarter.
- Get accounts and history transferred first. Ad accounts, pixel and conversion API access, the email platform, the creative library. Written into exit terms before you give notice.
- Freeze strategy changes during the handoff. Keep spend flat and let the incoming team learn the account first. Anyone restructuring your campaigns in week one is guessing.
- Move one channel at a time. Swapping paid media and email at once costs you the ability to tell what caused what.
- Set the judgment date before you start. Ninety days from the first full month, against the measure everyone agreed on.
Agency notice periods run 30 to 60 days, and some contracts hold your assets in ways you won’t enjoy discovering. Read that clause before signing the next one.
Common questions
Can a fractional CMO manage my agency? Yes, and that’s usually where the retainer earns out fastest. An operator with executive standing can rewrite scopes against outcomes, consolidate overlapping vendors, and hold a channel team to contribution margin instead of ROAS. Agencies perform better with someone competent above them, because the brief stops changing every three weeks.
Is an agency cheaper than a fractional CMO? Per retainer, roughly the same. Most brands run two or three vendors, so the agency layer costs $15,000 to $25,000 a month combined, and none of it buys anyone accountable for the total. Compare the full structure rather than one line item.
Do I need both? At $3M to $15M, usually yes. The fractional operator sets direction and owns the number, agencies supply throughput, and one or two in-house owners hold the work that has to stay inside. Very few brands in that band can afford enough senior in-house talent to skip the agency layer.
When should I hire in-house instead? When the function is continuous, proprietary, and you can absorb four to six months of search plus ninety days of ramp. Retention lifecycle, brand, and customer insight are the usual candidates. Also hire in-house once your fractional operator has built the function and the seat has become full-time work, a transition worth planning for from the start.
What if my agency says they do strategy too? Ask what number they own and what they’d cut. A channel agency’s strategy is channel strategy, which is real and valuable inside their box. The test is whether they’d recommend moving budget out of their own scope. Very few will, and that’s a structural conflict rather than a character flaw.
If you’re between $3M and $15M with competent vendors, no shared plan, and every decision routing through you, that’s an ownership bottleneck and the one I get hired to fix. Start with the fractional CMO overview if you’re still scoping the role. Here’s how engagements work when you’re ready to talk structure.