Three different jobs, routinely confused. The expensive mistake is hiring one and expecting all three.
The question is usually framed as a budget decision. It is really a decision about what kind of financial judgment the business needs and how often it needs it.
A bookkeeper records what happened. Transactions coded, accounts reconciled, bills paid, invoices sent. Accurate history, reliably maintained.
A controller owns whether the history is right. Close process, internal controls, accounting policy, the judgment calls about how something should be recorded. A controller catches the error a bookkeeper makes and prevents the next one.
A CFO owns what the numbers mean for what you do next. Forecasting, capital structure, pricing strategy, board and lender relationships, transactions. A CFO reads the same financials and tells you which decision they support.
These are three different jobs. The common and expensive mistake is hiring one and expecting all three.
A company at eight million in revenue has a bookkeeper who has been there five years and is genuinely good at the job. The founder is doing the controller work at night and the CFO work not at all. Nobody is wrong and nothing is failing, but pricing has not been revisited in two years, cash is managed by watching the balance, and the last board meeting produced a request for numbers nobody has produced yet.
Hiring a full-time CFO at that stage is usually premature. Promoting the bookkeeper into controller work usually fails, not for lack of ability but for lack of the specific experience the role requires.
These are not rules, but they hold more often than not.
Below two million in revenue, a competent bookkeeper and a good tax preparer is often the right answer, and we say so when it is.
Between two and ten million, most companies need controller-level oversight and periodic CFO input. Full-time hires for either are hard to justify and hard to attract.
Above ten million, or at any revenue with a transaction, an audit, or institutional investors involved, CFO-level judgment stops being optional.
The argument for fractional is not primarily cost, though it is usually cheaper. It is that you can buy the seniority you need at the frequency you need it. Two days a month of genuine CFO judgment beats a full-time hire who is over-qualified for most of what lands on their desk and under-qualified for the rest.
The argument against it is real too: a fractional team is not in your building, does not absorb culture by osmosis, and requires you to be deliberate about communication. Companies that need someone physically present in daily operations should hire.
Ask what decision you are currently making badly, or not at all, because nobody senior is looking at the numbers. If the answer is about accuracy, you need a controller. If it is about direction, you need a CFO. If it is both, you need a structure rather than a hire.
Eight questions on how your books run today. You get a score, a read on where you are fragile, and the order we would fix things in.
