On being the one number in the building that everyone else must trust
The board meeting is running long. The CEO has just finished the growth story, and everyone in the room is nodding along. Then someone asks the CFO the only question that matters: “Does the cash support this?” The room goes quiet in a specific way. Everyone else in that meeting is allowed to be optimistic. The CFO is not.
I have spent a career advising companies from the garage to the global stage, and I have sat across the table from a great many CFOs along the way, at companies raising their first institutional round and at companies six months from an IPO. The good ones share a quality that is hard to name and easy to recognize once you have seen it. While the rest of the company can afford to believe things, the CFO must know them.
Everyone else has a story. The CFO has the number.
A company runs on narrative. The CEO sells a vision, the head of sales sells a quarter, the head of product sells a roadmap that will, this time, ship on schedule. None of that is dishonest. It is how ambitious people build things that do not yet exist. The CFO’s job is to hold the one version of the company that must be true regardless of who is in the room. It is not the version that gets the round done or keeps the team’s morale up through a hard quarter, though a good CFO cares about both of those things. It is the version in which the numbers tie out, the cash lasts if it is supposed to, and the story the board hears on Tuesday still holds up when the auditors show up in March. That is a strange position to sit in. You are embedded in the enthusiasm and structurally obligated not to fully share it.
Why the job is quietly adversarial
Nobody sets out to make the CFO’s job hard. But look at where the incentives point. The sales team is measured on bookings, not on whether the contract terms will pass revenue recognition. The head of engineering is measured on shipping, not on whether the capitalized software hours were tracked correctly. The CEO is measured on the story investors believe, not on whether every assumption behind it survives diligence. The CFO is the only person in the building measured on whether all of it is true at once. That creates a particular kind of loneliness, quieter than the CEO’s, but real in its own way. You are the person who must tell the head of sales that the deal cannot be booked this quarter. You are the person who tells the board, three weeks before the round closes, that the runway is shorter than the deck says. Nobody enjoys being the correction. The good CFOs learn to deliver it so often, and so early, that it stops feeling like bad news and starts feeling like weather.
Know the number cold
Every great CFO I have worked with can answer, without opening a laptop, how many months of runway the company has, what the burn was last month, and what changes that math in either direction. That sounds like a low bar. It is not. Most finance leaders can find that number. The good ones carry it. Carrying it means the number changes how you show up in every other conversation. It means knowing, before the head of engineering asks, that the new hires push the runway conversation up by a quarter. It means that, when the board asks a hard question, you are not doing math in real time; you are reporting a fact you already knew was coming.
Be the translator, not just the reporter
A CFO who only reports the numbers is doing half the job. Boards and investors do not actually want the trial balance. They want to know what it means and what to do about it. The CFO who can turn twenty-two tabs of a financial model into three sentences a director can repeat to their partners the next morning is worth an entire finance function. The same skill runs the other direction. A great CFO translates the company’s ambitions into what the balance sheet can support and does it early enough that it shapes the plan rather than autopsies it. That is a very different job from producing the actuals a week after the quarter closes.
Say the hard thing before you are asked
The CFOs who get companies into real trouble are rarely dishonest. They are late. They wait for more certainty, or a better quarter, or a moment that feels less awkward, and the gap between when they knew and when they said something becomes the actual crisis. The good ones say it early and say it plainly, even when it is unwelcome, even when the CEO is mid-pitch to a new investor. A board that hears bad news in real time from its CFO will forgive almost anything. A board that finds out three weeks late forgives very little, and it remembers.
Build the systems before you need them
Nobody notices the close process, the revenue recognition policy, or the cap table hygiene until the moment they are under diligence, and by then it is too late to build them properly. The unglamorous work—a clean chart of accounts, documented internal controls, and a current 409A valuation when the options were granted—is what allows a company to survive a term sheet, an audit, or an acquisition without a surprise showing up in someone else’s spreadsheet. The CFOs who do this well treat every close like it might be read by a stranger someday, because eventually it will be—by an auditor, an acquirer’s finance team, or a plaintiff’s expert. Build for that reader from the first month, not the month you find out they are coming. The boring documents are never boring. Someone always reads them eventually.
Choose your outside team like you will need them in a crisis, because you will
A great CFO does not wait until the audit is difficult to have a real relationship with the audit partner. They have already talked through the structure with counsel before the term sheet is on the table. They do not treat the accounting firm, the bank, and the law firm as vendors to be managed, but as the small group of people who will be in the room during the worst week the company ever has. The cheap version of any of these relationships is expensive later. A banking relationship that only exists on paper will not answer the phone during a liquidity crunch. An audit firm chosen on price alone will not have earned the credibility to be believed by a skeptical acquirer. Build the relationships before the year you need them, because that is not the year to be building them. I say all of this knowing full well that I am one of those phone calls, so discount for self-interest accordingly. The pattern still holds.
Protect the close calendar like it is the product
The most concentrated risk in a finance organization is a CFO and a controller who are too busy to look up. Protect the monthly close the way an engineering team protects a release schedule, because a close that slips is not just an inconvenience; it is a company that does not actually know where it stands. The same discipline applies to the CFO personally. The best financial leaders I know guard blocks of real, undistracted time the way they guard the bank balance, because the analysis that catches a problem before the board meeting does not happen in the fifteen minutes between two calls.
The trust is the whole job
Strip away the models, the close calendars, and the term sheets, and what is left is trust: the board’s trust that the number is real, the CEO’s trust that the CFO will tell them the truth before they need to hear it, and the company’s trust that payroll clears on the fifteenth without fail. That trust is built slowly, in a hundred small moments nobody else in the building sees, and it can be spent in a single quarter of surprises. The CFOs who last understand that the balance sheet is not actually the asset they are protecting. Their own credibility is, and once it is gone, no model rebuilds it. The board meeting eventually ends. The room clears out, still nodding along to the story. The CFO stays a few minutes longer, closing the laptop last, because someone in that room must know, not believe, that the number holds. That part of the job does not change, no matter how large the company gets.
Find a peer group
No seat in the company has widened faster than the CFO’s, and AI is the reason. Finance leaders who once owned reporting and controls now sit inside strategy, systems modernization, risk, and capital allocation. They are the ones left holding the questions every AI project eventually reaches. What is the return? What do we fund, and what do we quietly stop funding? How do we measure productivity instead of guessing at it? How do we explain any of it to the board? Those are not questions a CFO wants to reason through in public, and no vendor is positioned to answer them honestly. Having watched this from the board side for years, I can tell you that AI transformation is not real until the CFO understands it, funds it, measures it, and helps govern it. That is a far easier job with trusted peers than without them.
That is why I joined the board of the CFO Executive Forum, part of Open Future Forum in Silicon Valley. The Forum convenes CFOs and senior finance leaders in off-the-record dinners and roundtables that are peer-first, not sponsor-first: partners are in the room when they add value, not to pitch, and the filter for admission is not whether someone can pay but whether they will make the room better. If you are a finance leader working through these decisions, I invite you to join us at an upcoming gathering. You can apply directly or learn more about the Forum here.
