Hiring a chief financial officer goes wrong long before anyone opens a résumé. It goes wrong the week the CEO and the board agree they need a strong CFO without agreeing on what the person is being hired to do. Everything after that — the specification, the slate, the interviews, the offer — inherits that ambiguity and amplifies it.
What follows is not a framework I invented. It is the set of questions I am actually asked by CEOs, board chairs, and CHROs at companies between $50 million and $10 billion in revenue, roughly in the order they come up, with the answers I give them.
We agree we need a CFO. What is the first decision?
Not the search. The mandate. The CFO title covers at least three distinct jobs, and the person who is genuinely excellent at one is frequently mediocre at the next. Before anyone drafts a position specification, the CEO and the board should be able to answer a single question in one sentence: what has to be true about this company's finance function in twenty-four months that is not true today?
The honest answer almost always sorts into one of three mandates.
The control mandate
You are hiring because the numbers are late, contested, or wrong. The symptoms are specific and unglamorous: a close that runs past business day fifteen; three charts of accounts left over from acquisitions nobody integrated; a revenue recognition policy under ASC 606 that the auditors have re-litigated two years running; a lease population under ASC 842 that still lives in a spreadsheet owned by one person; an ERP conversion that stalled at go-live and never recovered. This CFO's first year is close calendars, reconciliation discipline, a technical accounting hire, and a controller who can genuinely own the ledger. Candidates who thrive here have usually lived through a restatement, a material weakness remediation, or a first audit. Ask them about it — they are never shy.
The growth mandate
The numbers are fine. The decisions are not. This is the CFO you hire because the company cannot answer which customers, products, plants, or channels actually make money. The work is FP&A construction, pricing, and capital allocation. In SaaS, that means net revenue retention decomposed by cohort, CAC payback that survives contact with the sales comp plan, and a Rule of 40 position the CFO can defend in a board meeting without a slide. In industrial and manufacturing businesses, it means standard costs re-rolled on a real cadence, absorption variance explained rather than plugged, and make-versus-buy analysis the plant managers do not quietly ignore. This person is a business partner who happens to own accounting, and the smart ones hire a strong controller precisely because ledger mechanics are not their edge.
The capital-markets mandate
You intend to raise, sell, refinance, or list inside roughly thirty-six months. The job is the equity story, the quality-of-earnings defense, the bankers, the lenders, and the covenant package. If a listing is the destination, the terrain is concrete rather than conceptual: emerging growth company status under the JOBS Act runs five years from listing with an SEC inflation-adjusted revenue ceiling of $1.235 billion, and smaller reporting companies under $100 million in revenue are exempt from Section 404(b) auditor attestation under the SEC's 2020 accelerated filer amendments. Those thresholds decide what gets built, in what order, and at what cost. And from the first filed report onward, this person certifies personally under Sarbanes-Oxley Sections 302 and 906 — where a willful false certification carries penalties up to $5 million and twenty years. That is not a résumé line. It is a risk a candidate either has taken before or has not.
Can one CFO do all three? Rarely, and never simultaneously. Most CFOs have one dominant mandate and one credible adjacency. The practical move is to name the dominant mandate for the next twenty-four months, name the adjacency you will need after that, and accept that you are buying the first and renting optionality on the second. Companies that refuse to choose end up with a generalist who is adequate at everything the board is anxious about and excellent at nothing it actually needs.
Can we just promote our controller?
Sometimes — and more often than search firms like to admit. The promotion works when the mandate is control, and when the controller has already been operating a level above their title without being asked to. The diagnostic I use is blunt: has this person ever told the CEO no, with a number behind it, and made it stick? Controllers who have done that are running a function. Controllers who have not are running a department.
Where the internal promotion reliably fails is the capital-markets mandate. A first-time CFO learning S-1 mechanics, auditor negotiation, and banker management on live ammunition is an expensive way to discover the gap. It also fails when the real problem is that the CEO wants a thought partner and the controller's professional identity is accuracy. Accuracy is a virtue. It is not a strategy.
One more trap: VP of Finance is not CFO minus a title. The VP Finance role is typically inward and analytical; the CFO role is outward and accountable. Promoting across that line without deliberate scaffolding — a board mentor, a strong technical accounting deputy, a defined first-year mandate — is how a good operator gets broken in public. If you are weighing this, think about the whole structure rather than the one seat; I wrote about that tradeoff in more depth in building a finance leadership team.
Does the CFO have to be a CPA? Does Big Four matter?
Short answer: the CPA matters for the control and capital-markets mandates and matters very little for the growth mandate. If your CFO will sign certifications, negotiate with an audit committee, and defend technical positions, the credential is a floor, not a differentiator. If your CFO's job is pricing, capital allocation, and unit economics, requiring a CPA screens out a large share of the best candidates for no defensible reason.
Boards frequently write a CPA requirement into a CFO specification because someone conflated it with Item 407(d)(5) of Regulation S-K — the audit committee financial expert. That is a director designation. It belongs on the board, not on the executive team, and satisfying it is the nominating committee's problem, not the CFO search's.
Big Four training is a proxy, and like all proxies it decays. What I actually probe for instead: Has this person survived a PCAOB-inspected audit as the client? Have they owned an ERP conversion end to end, including the part where the first close after cutover goes badly? Have they built a technical accounting function, or only inherited one? Those questions separate candidates. Firm logos do not.
What belongs on the scorecard — and what should never?
Scorecards fail because they list attributes instead of outcomes. Attributes are unfalsifiable; outcomes have dates. Five to seven outcomes, each with a deadline and an owner-visible measure, is the whole document.
What a real line looks like: By month nine, standard costs are re-rolled across all four plants and absorption variance is explained monthly to within two points of plan. Or: By month six, the close is complete by business day eight with a documented reconciliation package the auditors accept without a management letter comment. Or: By month twelve, a debt-to-EBITDA covenant package is in place with at least 1.0 turn of headroom under the downside case.
What should never appear:
- Years of experience. It measures elapsed time, not reps. A CFO with eight years across two turnarounds has more relevant reps than one with twenty years of steady state.
- Industry as a hard gate. Industry is a proxy for a specific competency. Name the competency instead. "Must come from manufacturing" is lazy; "must have owned a multi-plant standard costing environment" is a real requirement that a candidate from an adjacent sector may fully satisfy.
- Soft-skill adjectives. "Strong communicator" and "executive presence" are what interviewers write down when they cannot articulate what they saw. They also do most of the quiet demographic filtering in executive hiring.
- School. There is no evidence it predicts CFO performance. There is considerable evidence it narrows the slate.
How do we interview for the mandate instead of listening to a career recital?
Most CFO interviews are a candidate narrating their résumé backward while four executives nod. The fix is structure plus one genuinely technical probe matched to your mandate.
Structure first: walk the career chronologically, role by role, and at each stop ask the same four questions — what were you hired to do, what did you inherit, what did you actually change, and who can confirm it. The consistency is the point. Candidates who are embellishing tend to survive question one and come apart at question three.
Then go technical, and go specific to your lane:
- SaaS or subscription: walk me through your ASC 606 standalone selling price allocation for a multi-year contract with a ramp and a mid-term upsell. Then tell me how you handled the swing back to immediate domestic research expensing under the new Section 174A after three years of capitalizing under Section 174 — and what it did to your cash tax forecast.
- Mining, minerals, or rare earths: walk me through your last reserve statement under SEC Regulation S-K Item 1300, how you reconciled the Qualified Person's technical report summary to the carrying values, and where you and the QP disagreed.
- Industrial and manufacturing: walk me through your Section 45X credit position and whether you transferred the credits for cash under Section 6418 or carried them. Then explain your last inventory reserve methodology change and how you sold it to the auditors.
- Any group above roughly €750 million: tell me how you scoped Pillar Two top-up tax and what surprised you.
A CFO who has genuinely done the work answers these in specifics, including the parts that went badly. A CFO who has supervised someone doing the work answers in principles. Both can be right hires — but you should know which one you are getting.
What are references supposed to surface that interviews cannot?
On-list references confirm employment. That is nearly all they do. The reference work that changes decisions is off-list and sourced from the market, not from the candidate.
For a CFO specifically, the four calls that matter most are the audit partner or engagement lead, the lead banker or lender on their last financing, the CEO they reported to two roles ago (not the current one, who has incentives), and two people who reported to them and subsequently left. That last pair is the highest-yield call in the entire process and the one almost nobody makes.
The question that produces signal is never "was this person good?" It is "what did this person need around them to succeed, and what happened when they did not have it?" Every CFO has a dependency. The good references name it in one sentence. This is also the part of the process where an intermediary earns their keep, because a sitting CFO's network will speak candidly to a third party in a way they never will to a prospective employer.
How long does a CFO search actually take?
From an honest kickoff to a signed offer, plan on twelve to sixteen weeks, then add the notice period — sixty to ninety days is common at this level, and garden leave can extend it further. The phases: two to three weeks of calibration and mandate alignment, four to six weeks of market mapping and approach, a first real slate somewhere around week six to eight, and then interviews, references, and negotiation.
What extends it, in order of frequency: an unresolved mandate that surfaces in week seven when two board members disagree in front of a finalist; a compensation philosophy that was never agreed internally, particularly the equity component; relocation the company assumed and the candidate never accepted; and a CEO who keeps meeting "one more" candidate because none of them are the person they imagined.
One timing constraint that catches public and pre-IPO boards off guard: Form 8-K Item 5.02 requires disclosure of a CFO departure within four business days. If you are replacing a sitting CFO who does not yet know, the entire search is timed backward from the resignation date, runs confidentially, and cannot use a public posting. That is a materially different process, and it is one of the clearest cases for a retained engagement — I unpacked the mechanics in how retained executive search actually works.
What kills the hire in year one?
Rarely competence. Almost always one of these four.
- Mandate drift. You hired a control CFO in January; a sale process appeared in June. The person is not failing — the job changed and nobody said so out loud.
- The two-CFO problem. The founder or CEO still owns the model, still sets pricing, still talks to the lender. The new CFO is accountable for outcomes they do not control and leaves inside eighteen months, usually politely.
- No data on day one. The CFO cannot get clean access to the systems for ninety days because IT is mid-migration, so the first board meeting they own uses numbers they did not build and cannot defend.
- Comp built for the wrong mandate. A capital-markets CFO on a pure EBITDA bonus will optimize EBITDA, not the exit. The incentive design has to match the mandate you named in week one.
Each of these is preventable at the specification stage and nearly unrecoverable at month nine.
When do we run this internally, and when do we bring in a firm?
Run it internally when the mandate is control, the market is one you already know, your talent brand is strong enough that good people answer, and you have the internal bandwidth to work a slate for three months without dropping it. Plenty of companies do this well and should.
Bring in a firm when the search is confidential, when the candidates you need are employed and not looking, when it is a first public-company CFO and you need someone who can assess certification-level experience credibly, or when the board wants a defensible process on record. The criteria for choosing between firms are worth more scrutiny than most boards give them — I laid out the evaluation questions in how to choose an executive search firm, and the specific triggers for a CFO seat in when to bring in a CFO search firm.
Either way, the sequencing is the same and the order is not negotiable: mandate, scorecard, slate, structured interviews, off-list references, offer designed to the mandate. Skip the first step and the other five are decoration. That is the discipline we hold to at Turnkey Recruiting, and it is the same discipline a good internal team can hold to without us. For a tighter operational walkthrough of the same sequence, see the board-ready playbook on hiring a CFO.
Questions boards ask at the end
These are the ones that come up after the mandate conversation, almost every time.
Frequently asked questions
Should the CFO report to the CEO or the board?
To the CEO, with an explicit and protected line to the audit committee. The CFO's independence on financial reporting is not a courtesy — it is the structural reason the audit committee exists. If a candidate asks about the audit committee's access to them without a chair present, that is a good sign, not a red flag.
How much should we weight industry experience versus stage experience?
Stage, in most cases. A CFO who has taken a company from $150 million to $600 million in an adjacent sector will usually outperform a same-industry CFO who has only operated at steady state. The exceptions are industries where the accounting itself is the job — mining reserve reporting under S-K 1300, regulated utilities, insurance. There, industry experience is a genuine gate rather than a proxy.
Our CFO is leaving in six weeks. Do we need an interim?
If the close, the audit, or a financing sits inside the gap, yes. Name an interim quickly, name them publicly, and tell them explicitly whether they are a candidate for the permanent seat. An interim who thinks they are auditioning will optimize for looking good rather than for a clean handoff, and the incoming CFO inherits the consequences.
How do we assess a first-time CFO candidate fairly?
Look for scope that exceeded the title. A VP of Finance who ran the lender relationship, owned the board deck, and made the call on a contested revenue position has already been doing the job. Then be honest about what the company can supply: a first-time CFO needs a strong controller, a board mentor, and a mandate that does not include an S-1 in year one.
What is the single best predictor of CFO success in the first two years?
Alignment between the mandate you named and the mandate the candidate has actually run before. Not credentials, not tenure, not industry. When a CFO fails inside two years, the post-mortem almost always shows the company hired for one mandate and needed another — and knew it before the offer went out.