The short version: To vet a fractional CMO, run six layers instead of one interview: check portfolio proof, verify the metrics they claim, contact 3–4 references you reach yourself (not ones curated on a call), consider a paid test sprint, write replacement clauses into the contract, and set 30-day checkpoints. The US Department of Labor puts a bad hire at 30% of first-year pay, so a week of vetting is cheap insurance. O-CMO does this before any profile goes live: tokenized reference forms sent straight to former clients and peers, scored across strategy, execution, and communication, with matching in 5–7 business days.
Disclosure: O-CMO is our marketplace. We describe our own verification method below as the worked example, then show you how to run the same rigor yourself if you hire directly. The external sources are linked so you can check them.
You have a shortlist. Three or four fractional CMOs, all with good decks and a story about tripling pipeline somewhere. The problem is that a polished storyteller and a real operator look identical for the first 45 minutes. The gap shows up in month three, when the person who “led growth” turns out to have been one of nine people in the room. Vetting is how you find that gap before you sign.
This guide is about the vetting process. If you want the questions to ask in the interview itself, we cover those separately in fractional CMO interview questions. Here we assume the conversations went well and now you need to confirm the story is true.
Why Vetting a Fractional CMO Is Different
A fractional CMO sets your strategy and spends your budget on partial hours, often across several clients. You do not observe them daily, so upfront verification matters more than it does for a full-time hire you can course-correct in standups.
The base rates are not reassuring. In a ResumeBuilder survey, about one in three Americans (32%) admitted to lying on a resume, and marketing claims are unusually easy to inflate because so much of the work is attributed to a team. Meanwhile the US Department of Labor figure most recruiters cite puts the cost of a bad hire at about 30% of the person’s first-year earnings, before the lost quarter of go-to-market time. For a leadership role that is conservative. Vetting is the cheapest lever you have.
The 6 Layers of Vetting a Fractional CMO
One interview tests how someone presents. These six layers test whether the presentation is true. Run them in order; each one is cheap until the last, and most candidates who are going to fail fail early.
Layer 1: portfolio proof, not portfolio claims
Ask for two or three engagements they will talk about in detail, then push past the headline. What was the situation when they arrived? What did they decide, and what did they decide against? What would they do differently? A real operator gives you texture and tradeoffs. A storyteller gives you a case study that sounds like the deck, because it is the deck. If every example ends in a clean win, that is the flag.
Layer 2: verify the metrics they claim
“Grew revenue 3x” is a claim, not a fact, until you know the base, the timeframe, and the person’s actual role. Ask for the starting number and the window. Ask who owned the budget and who else was in the room. Then confirm it with a reference, because the metric and the attribution are exactly what an inflated story gets wrong. Solo credit for a team outcome is the single most common exaggeration in marketing.
Layer 3: reference checks, done right
This is the layer most buyers do worst, so it gets its own section below. The short version: the references a candidate reads to you on a call are the ones who will say nice things. The signal is in the references you reach yourself, and in the questions you ask them.
Layer 4: the paid test sprint
For senior hires, the most reliable signal is a small piece of real work. Pay for a two to four week diagnostic: a 30-day plan, a channel audit, or a positioning teardown of your actual product. You get two things. The deliverable, useful either way, and the evidence of how they think when the answer is not rehearsed. Pay market rate for it. Free test work attracts the wrong people and tells you nothing about how they behave once money is on the table.
Layer 5: contract protections
Vetting continues after signature. Structure the engagement so a wrong fit is cheap to exit: month-to-month terms after any initial period, a 30-day notice window, and a defined first-month scope you can judge. A replacement or refund guarantee for month one shifts real risk off your side of the table. If a provider will not put any of this in writing, that answer is itself a data point.
Layer 6: the 30-day checkpoints
Agree upfront on what “on track” looks like at day 30, day 60, and day 90. Not revenue, which is too slow to judge a fractional leader in a month, but leading signals: a written strategy, a prioritized plan, the right questions asked about your data, early experiments running. The checkpoint is not a performance review. It is the last layer of vetting, run in the open, where a mismatch is still cheap to fix.
How to Run Reference Checks That Actually Tell You Something
Here is how O-CMO does it, because our verification method is the same method we would tell you to run yourself. Every CMO in the catalog is reference-checked before their profile goes live. We do not take the references on a call. We send tokenized reference forms directly to named former clients and peers, so the feedback comes back to us, not filtered through the candidate. Each response is scored across three axes: strategy, execution, and communication. The scores drive a tier assignment, and the whole cycle runs in 5–7 business days. A candidate who cannot produce reachable former clients does not get published.
You can run the same rigor when hiring directly. Three principles carry most of the weight.
- Reach the references yourself. The names a candidate hands you are pre-warmed. Ask for former clients specifically, then confirm the relationship on LinkedIn before you call. Where it is appropriate, a back-channel reference through a mutual contact you already trust tells you more than any list the candidate curates. SHRM’s guidance on reference checks covers how to do this without crossing legal or ethical lines.
- Ask behavioral questions, not character questions. “Was she good?” gets you a yes. “What did she own versus the team?” and “What would you have wanted her to do differently?” get you the truth. The best single question: “Would you hire this person again for the same role, and for what?” Hesitation is the answer.
- Cross-validate, never accept in isolation. One glowing reference is a data point. The same story from three independent people is a fact. Contradictions between references and the candidate’s own account are where the gap lives.
10 Red Flags When Vetting a Fractional CMO
These are drawn from real screening. None is fatal on its own. Two or three together is a pattern.
- They cannot name a failure. A senior operator has scars. Someone whose every story is a win has either never taken a real risk or is editing.
- They claim solo credit for team outcomes. “I grew it 4x” with no mention of the team, the founder, or the budget owner. Marketing wins are almost never solo.
- No questions about your data. A real CMO wants to see your funnel, your CAC, your retention curve before quoting a plan. Someone who pitches a strategy without asking is selling a template.
- Only peer references, no clients. Peers vouch for likeability. Former clients vouch for results. If nobody who paid them will get on a call, ask why.
- Metrics with no base or timeframe. “3x growth” that dissolves when you ask “from what, over how long?” was never a number.
- Vague on their actual hours. A fractional leader juggling clients should tell you exactly how many hours you get and when. Hand-waving here becomes a scheduling problem later.
- Channel-first, strategy-never. If they open with “we’ll run paid and SEO” before understanding your business, you hired a specialist, not a leader.
- They resist any test or trial. Confidence welcomes a paid diagnostic. Resistance to proving the work is a tell.
- Won’t sign fair exit terms. A month-one guarantee or 30-day notice is standard. Pushback suggests they expect you might want out.
- The story shifts between the deck and the reference. Small contradictions on scope, budget, or role are the loose thread. Pull it.
Your Fractional CMO Verification Checklist
One pass through each layer, with what you are actually confirming and the effort it takes.
| Layer | What you verify | Effort |
|---|---|---|
| Portfolio proof | Real decisions and tradeoffs behind 2–3 engagements | 1 call |
| Metrics verification | Base, timeframe, and the candidate’s actual role | 30 min + reference |
| Reference checks | 3–4 references you reach, at least 2 former clients | 2–3 calls |
| Paid test sprint | How they think on your real problem | 2–4 weeks, paid |
| Contract protections | Month-to-month, 30-day notice, month-1 guarantee | Legal review |
| 30-day checkpoints | Strategy, plan, right questions, early experiments | Ongoing |
If that looks like a lot of process, it is. This is why marketplaces exist. On O-CMO, layers one through three are done before a profile is published, and the contract terms in layers five and six come standard: matching in 5–7 business days, a month-one replacement guarantee, month-to-month after with 30-day notice, and account management on the engagement. You still run your own interview and, if you want, your own test sprint. The verification floor is just already set.
Frequently Asked Questions
How many references should I check for a fractional CMO?
Three to four, and at least two should be former clients rather than peers. One reference is a data point; three telling the same story is a fact. The value comes from reaching people yourself and cross-validating, not from the raw count. Beyond four you tend to hear the same signal repeated.
What should I ask a fractional CMO’s reference?
Ask behavioral, specific questions. What did this person own versus the team? What would you have wanted them to do differently? Would you hire them again for the same role? Avoid “were they good,” which only ever returns yes. Hesitation on the rehire question is usually the real answer you came for.
Should I pay for a test project before hiring?
For a senior fractional CMO, a paid two to four week diagnostic is the most reliable signal you can buy. You get a useful deliverable and see how they think on your real problem. Always pay market rate. Free test work attracts the wrong candidates and tells you nothing about how they operate once money is involved.
What if all their references are peers, not clients?
Treat it as a flag worth a direct question. Peers speak to how someone works; clients speak to whether the work produced results. If no former client will get on a call, ask why, and lean harder on a paid test sprint. For early-career fractional CMOs some peer weighting is normal, but the balance should shift toward clients as they gain track record.
The Bottom Line
A good interview tells you how someone talks about their work. Vetting tells you whether the work happened the way they say. Run the six layers, reach the references yourself, and pay for a small piece of real work before you commit to the big one. The storytellers wash out somewhere between the metrics check and the reference call, which is exactly where you want them to. If you would rather start from a shortlist where the first three layers are already done, that is what a verified marketplace is for: see whether you should hire a fractional CMO at all, then browse reference-checked profiles with the vetting floor already set.
Methodology: O-CMO verification process described from our own practice (O-CMO marketplace data, 2026). Third-party figures on bad-hire cost and resume misrepresentation attributed to the US Department of Labor and a ResumeBuilder survey respectively; reference-check guidance sourced from SHRM. Sources linked inline, verified October 2026.
