Most people weighing fractional CFO work already have the skill. You have closed the books, defended a model to a board, and talked a founder off a ledge during a bad quarter. The question that actually keeps you in the salaried seat is quieter and more practical: can you replace one steady paycheck with a handful of part-time clients, and stay solvent during the months it takes to build that roster? That is the real subject here. Not a list of finance skills you already have.
The math that decides whether you can leave
Start with the number, because everything else follows from it. A full-time CFO at a growth-stage startup earns somewhere around $220k to $300k in salary. A fractional retainer for a few days a month typically runs $4k to $10k, depending on how much you own. So replacing a senior salary takes roughly three to five steady clients, and you almost never sign them in the same month.
Run it forward. Five clients at $6k each is $360k a year gross, which sounds like a raise until you subtract self-employment tax, your own health insurance, unpaid vacation, and the occasional gap between engagements. Net, that stack lands close to the salary you left. The catch is the word steady. Getting from your first client to a full roster is the part that takes real time, and it is where most of the risk lives.
The people who make the jump cleanly tend to do one of two things first: bank six to nine months of personal expenses, or line up one anchor client who signs before they hand in notice. Doing neither, and quitting into an empty pipeline, is how a good CFO ends up taking the first lowball retainer out of fear.
You bill for an outcome, not a day rate
The instinct from a career of salaried work is to price yourself by the hour, like a contractor. Resist it. Founders are not buying hours of finance. They are buying a raise that closes, a runway they can trust, a board deck that survives diligence. Price the retainer against that outcome. A CFO who can carry a Series A model through to a signed term sheet is worth far more than the same hours billed as bookkeeping oversight.
Vague pricing invites scope creep, which is the quiet tax on fractional work. Write down what the retainer covers, say a monthly close review, a rolling forecast, and one board deck a quarter. Then name what sits outside it and bills separately, like a fundraise sprint or audit support. Founders respect the line once it is drawn. They resent discovering it mid-engagement.
The cash-flow trap of your first year
Salaried income is flat. Fractional income is lumpy, and the lumpiness is the thing new fractional CFOs underestimate most. One client wraps a raise and offboards. Another pauses to conserve cash. If a single client is half your revenue, losing them is a fifty percent pay cut with two weeks notice.
Concentration is the danger, and a warm pipeline is the only real hedge. The CFOs who get through year one keep talking to prospects even when they are fully booked, so a churned client is replaced in a few weeks instead of a hollow quarter. This is exactly why the work of finding clients ends up mattering more than the finance itself, and why it deserves a real system rather than luck.
Where fractional CFO work actually originates
Founders rarely open a marketplace to shop for a CFO. Money problems surface in conversation, to people already close to the business. The lawyer running the round hears it. The recruiter watching headcount outrun the budget hears it. Another founder who just survived the same crunch hears it. The introduction goes to whoever those people already trust with something this sensitive.
That is the logic behind a referral circle: a small group of non-competing independent professionals, say a startup lawyer, a technical recruiter, a fractional COO, and a growth marketer, who all serve the same kind of company and pass work to each other. When one of them watches a client head into a raise with no finance leadership, they hand the founder a name instead of a search. Referna builds those circles, groups of eight to twelve peers who each interview and share a real client story before joining, moderated by Gordy, our AI community manager. The point is simple: the client who needs you should hear your name from someone they trust, before they ever start looking.