In house marketing vs agency is the wrong question

Every CMO I talk to eventually asks the same question in different words: in house marketing vs agency, which one actually works. The honest answer is that the question is framed wrong. Standard agency selection runs on a pitch, a case study deck, references, and a scope negotiation. Standard in house hiring runs on a resume, two structured interviews, and a reference check. Neither process evaluates the thing that determines the outcome, which is the judgement of the specific person who ends up changing bids and writing copy on a Tuesday afternoon.

The gap between the pitch team and the delivery team is the oldest problem in professional services, and it persists because it is commercially rational. Senior people win work. Junior people are how the work stays profitable. In house hiring has its own version: the person who interviews well is not always the person who can run a channel mix under budget pressure without losing their nerve.

What the standard process actually measures

Four questions that get past the pitch

1. 1

   Who is in this account every week, and what else do they run?

   You want names and a number. An operator carrying eight accounts is a coordinator with a login. Three or four is a working load. If the answer is vague, the answer is bad.

2. 2

   Tell me about an engagement or role that did not work

   The useful answer contains a specific diagnosis and a change in method. The useless answer blames the client's product, budget or internal politics. Everyone has failures; only some people have learned from them.

3. 3

   Look at our account for an hour and tell us what you would change

   Pay for the hour if you have to. Nothing in a capability deck or an interview loop approaches the signal you get from watching someone reason about your real data in real time. It also filters out anyone whose expertise lives in the pitch.

4. 4

   What would you refuse to do?

   An operator with a point of view has boundaries: channels they will not run, tactics they think are waste, targets they will not sign up to. A generalist will say they are flexible, which is another way of saying they have no opinion to defend.

Flexibility in a pitch or an interview usually means an absence of opinion. You are paying for the opinion.

A worked comparison: fractional operator vs junior in house hire

The structural signals, whichever way you go

• Layers. Every layer between you and the operator adds a day of latency and removes a degree of accountability, whether that layer is an account manager or a middle manager you inherited.
• Incentives. If an agency's fee scales with media spend, you have hired someone whose income increases when your budget does, whatever their intentions. If an in house hire's bonus is tied to lead volume, expect volume over quality.
• Contract or notice terms. A twelve month agency lock in with a three month notice period is a commercial structure designed for retention, not for results. A senior in house hire with no real onboarding plan is a slow bleed of the same kind.
• Named people in the agreement. If the operator you evaluated is not in the contract, you did not buy them. If the candidate you interviewed is not the one who shows up on day one, you did not hire them.

What this looks like across company stages

Early stage companies almost always underrate the fractional option because a full time hire feels like commitment and seriousness. A fractional lead at three days a week of attention, working alongside a specialist for execution, is usually the faster path to a repeatable channel before the company can properly evaluate a full time candidate anyway.

Companies past thirty million in revenue tend to make the opposite mistake. They keep the fractional arrangement going long after the program needs full time ownership, because the fractional lead is good and nobody wants to disrupt a working relationship. The tell is simple: if the operator is turning down your requests because of hours, not judgement, it is time to convert the role.

The strongest case for going in house anyway

Where fractional growth lead arrangements go wrong

The fractional model fails in one predictable way: the operator quietly takes on a fourth or fifth account and your twelve hours a week becomes eight. Nobody announces this. It shows up as slower replies and shallower answers in the working session. Ask directly, every quarter, how many active accounts they are carrying, and compare it to the number you were told at the start.

In house hiring fails in a mirrored way. The strong candidate you hired for strategic judgement gets pulled into execution because the team is short-staffed, and eighteen months later they are running Google Ads campaigns instead of setting the channel mix. If you hired for judgement, protect the hours that judgement needs, or you have quietly rehired a coordinator at a strategist's salary.

A short checklist before you sign anything

A second worked example: the account load math

Run the same arithmetic on any operator before you sign, agency or in house. Ask for the account count, subtract a fair estimate for internal overhead, and divide. If the resulting number is under six hours a week for a retainer that assumes strategic ownership, you are paying strategist rates for coordinator hours.

An operating protocol for the first ninety days

1. 1

   Week 1: confirm the account load in writing

   Get the named operator's current account count and weekly hour commitment in the kickoff email, not just the sales call. Verbal promises made during the pitch are not the same as a written commitment you can point back to in month four.

2. 2

   Weeks 2 to 4: watch response time before you watch output

   Output takes a quarter to show up. Response time to a hard question shows up immediately. If a direct question about strategy takes four days to get a real answer, that is your account load signal arriving early.

3. 3

   Day 30: run the working session test again

   The hour-with-your-data test is not a one time filter for the pitch stage. Run it again thirty days in, on live data instead of a sample, and see whether the sharpness from the pitch has survived contact with the actual account.

4. 4

   Day 60: name one thing that has not worked

   If the operator cannot name a test that failed by day sixty, either nothing risky has been tried or you are being managed rather than informed. Both are worth raising directly before the quarter closes.

5. 5

   Day 90: renegotiate hours or scope, not just price

   If the account load has crept, as it usually does, the ninety day review is the natural point to either add hours, cut scope, or accept a slower cadence. Doing this in the open protects the relationship more than quietly tolerating the drift.

The reference call that actually tells you something

Most reference calls ask whether the agency or candidate was good to work with. Ask instead what they refused to do, what they changed their mind about mid engagement, and how quickly they responded when something broke. Those three answers describe an operator. Praise for communication style describes a personality.

The cost of getting this wrong twice

Companies that hire on the pitch instead of the person do not usually fail once. They fail, blame the model rather than the diligence, switch from agency to in house or back again, and repeat the same untested selection process on the other side of the fence. The fix is not switching categories. It is fixing the test.

Run the same working session test on every candidate, agency or in house, before you commit budget or a salary line. It costs an afternoon. It is cheaper than a wrong hire by an order of magnitude, in either direction.

What I would do with a live decision this quarter

If I were choosing today, I would run three concurrent working sessions, one with an agency candidate, one with a fractional operator, one with an in house finalist, all against the same real account data, same questions, same hour. Whoever gives the sharpest, most specific answer about what they would change gets the role, regardless of which box they came from. I have made this call both ways and the working session has never once misled me.

The trade you are making

Hiring the operator, whether fractional or in house, has a real cost. Small senior teams have limited capacity, they will not absorb unlimited scope, and they cannot be staffed up in a week when your board asks for a new channel. A large agency or a bigger in house team genuinely solves for breadth and continuity.

But breadth is worth less than it sounds when the work is decided by judgement rather than by throughput. Most growth problems are not solved by more hands. They are solved by one person with context and authority making a decision quickly, and then making the next one. That person can sit inside your company or outside it. The org chart is the least important variable in the decision.