A traditional agency search, run with a consultant and a formal RFP, takes three to five months and can run close to a million dollars once pitch fees, staff time and consultant costs are totaled, according to the Association of National Advertisers. Most CEOs approving that spend have never run an agency search themselves and have no reliable way to know whether the shortlist in front of them is actually the best available or just the best at pitching.
That gap between who pays for the agency and who can actually judge one is the real argument for giving a fractional CMO final say. It is not about turf. It is about putting the decision in the hands of the person in the room who has seen enough agency work, good and bad, to tell the difference before the contract is signed rather than eighteen months into it.
A pitch is a controlled performance, not a sample of the work
The 4As and ANA’s own Cost of the Pitch research, a follow-up to their earlier Agency Search Simplification Report, exists because both organizations concluded the traditional review process was expensive enough to warrant its own study. A pitch team is usually the agency’s most senior talent, assembled specifically for the win, and frequently different from the team that will staff the account once it is signed. A CEO evaluating four such performances in a single week is comparing four best-case simulations, not four working relationships.
A fractional CMO who has actually managed agency relationships knows what to ask underneath the pitch: who specifically staffs the account day to day, what the agency’s last three client transitions looked like, how account strategy actually gets made once the pitch team goes back to their other clients. That is domain knowledge a CEO evaluating slides cannot easily replicate, no matter how sharp the CEO is in every other part of the business.
Relationships that start on a flawed process tend to end early
Agency relationships have grown shorter across the industry for years, and a process built to reward the best pitch rather than the best long-term fit is part of why. A search that optimizes for a persuasive hour in a conference room selects for agencies that are good at pitching, which is a real skill but a different one from being good at running an account for three years. When the mismatch surfaces, usually six to twelve months in, the company is back in another expensive search, having paid twice for a decision made once.
Setup’s 2024 summary of ANA agency-search data put the traditional process at three to five months before a client even reaches a signed contract. A fractional CMO with standing agency relationships and a clearer read on what good work looks like can frequently cut that timeline and, more importantly, cut the odds of ending up back in the same search a year later.
The CEO’s instinct is not wrong, it is just unverified
PwC research has found that 62% of executives rely more on intuition and personal experience than on data when making a decision, and KPMG’s research puts the share of executives who highly trust their own organization’s data at only 35%. Agency selection sits squarely inside that gap. A CEO’s gut read on which agency team felt sharpest in the room is a real signal, built from decades of judging people and pitches in other contexts, but it is not the same signal as knowing which agency actually delivers once the pitch team disperses.
Giving a fractional CMO final say does not replace the CEO’s judgment with data. It adds a second, calibrated judgment from someone whose gut read was built specifically on agency work, sitting alongside the CEO’s broader instinct rather than instead of it.
Final say is not the same as sole say
The honest version of this arrangement gives the fractional CMO the final technical call, the way a CFO gets final say on a financing structure, while the CEO retains veto power over anything that conflicts with strategy, budget, or a relationship the CEO has reasons to protect. That is a narrower grant of authority than it sounds, and it is also the version most CEOs are comfortable making once they understand it is not a blank check.
Where this breaks down is when a company hires a fractional CMO but never formally grants the authority, leaving the CMO to build a recommendation the CEO then quietly overrides for reasons never discussed. That version wastes the fractional CMO’s judgment without ever using it, and it is more common than either side likes to admit.
A pitch is a controlled performance by an agency’s best people. Final say belongs to whoever can tell that apart from the team who will actually staff the account.
The part nobody wants to hear
Most CEOs who resist giving a fractional CMO final say on agency selection are not protecting good judgment, they are protecting the habit of deciding everything themselves, built from years of running a company with no marketing leader at all. That habit made sense when there was nobody with more relevant judgment in the room. It stops making sense the moment the company hires someone specifically because that person’s judgment on this exact question is sharper.
The reverse failure is just as common. A fractional CMO who never asks for the authority explicitly, and instead hopes the recommendation will simply be accepted, has effectively volunteered to do the work of a decision without ever getting the decision. Naming the authority out loud, with its real limits attached, is uncomfortable for about five minutes and saves both sides a wasted search eighteen months later.
Write the authority down in the engagement letter
Final technical say on agency selection, CEO veto on strategy and budget conflicts. Two sentences, agreed before the first search starts.
Ask every finalist agency who actually staffs the account
Not who pitches it. A pitch team and an account team are frequently different people, and that gap is where relationships fail.
Call two former clients who left the agency, not just current ones
Current clients rarely volunteer the full picture. A former client, especially one who left recently, usually will.
Set a review date at ninety days, not a year
Catching a bad fit at ninety days costs a conversation. Catching it at month eighteen costs another full search.
The question is not whether the CEO’s instinct is good. It is whether the person in the room evaluating agencies has seen enough of them to know which instinct to trust.
Getting that authority structure right at the start of an engagement is usually one of the first things we put in writing, and how we help describes where it sits in the wider work.