By Andreas Pettersson, Founder, Leaders ADAPT
The owner of a $14 million distribution company told me he did not need a CFO because his controller was excellent. He was right about the controller. The books were clean, the close was on time, the tax filings never slipped. Then I asked him three questions. How much cash will you have on the first of next quarter. What is your margin by customer. What will the bank need to see before it extends the line in March. He answered all three with a version of the same sentence: I would have to work that out.
That is the whole question of when to hire a CFO. Not whether the numbers are right. Whether anyone is deciding with them.
I have held the seat myself. When I became the CEO of a Canon subsidiary at 37, I was CEO, CFO and CTO at once for a stretch, and I learned the hard way that a finance function has two jobs and most people in it are trained for one. Keeping score is the first. Telling the CEO what the score means before the quarter closes is the second. This page gives you seven signs that the second job is sitting empty in your company, what each one costs you, and the three ways to fill the seat.
Quick answer: Hire a CFO when the decisions have outgrown your controller: the owner forecasts cash, pricing is set by gut, the bank or a buyer asks for numbers you cannot produce, or a sale is inside two years. Most owner led companies between $5 million and $50 million do not need the seat full time. A fractional CFO one to three days a week, or an advisory CFO a few hours a week, fills it first.
Leaders ADAPT places fractional and advisory CFOs from its own bench and stays accountable for the engagement, priced as a fixed fee quoted after the first call. The market figures on this page come from named third parties, are linked, and carry the date I read them. If their number has changed, theirs wins. No provider named here pays us.
Controller vs CFO: the difference in one table
Most owners use the two titles as if the CFO were a senior controller. They are different jobs with different questions, and the table is the fastest way I know to show it.
| Question | Controller | CFO |
|---|---|---|
| What is the job | Record what happened, accurately and on time | Decide what happens next, with the numbers |
| Time horizon | Last month | Next 13 weeks and next three years |
| Core output | The close, the ledger, payroll, tax, compliance | Cash forecast, pricing and margin, financing, the plan |
| Who they talk to | The owner, the bookkeeper, the auditor | The owner, the bank, the board, the buyer |
| Measured on | Accuracy and timeliness | Whether the company made better decisions |
| Typical pay signal | Salaried, inside the finance team | Executive, sits on the leadership team |
A controller is paid to be accurate. A CFO is paid to be right about the future. Most owner led companies I meet have a good controller and an empty CFO seat the owner fills at night, badly, because the owner is also the CEO, the head of sales and the person the biggest customer calls.
The gap shows up in my own data. In the Leaders ADAPT coaching and assessment report for 2026, 57 percent of 84 owner CEOs raised cash, profitability and pricing as a live problem, the most common theme ahead of hiring, sales and strategy. Those are CFO questions, and the people asking them did not have a CFO.
The seven signs, what each one costs you, and the fix
Here is the table I use in a first call. If you recognise two of these, the seat is open. Four or more and it has been open for a while.
| Sign | What it costs you | The fix |
|---|---|---|
| 1. You build the cash forecast yourself, or check the bank balance instead | Owner hours, a line of credit drawn too late, a payroll scare | A 13 week cash forecast owned by finance, reviewed weekly |
| 2. Pricing and margin are set by gut | Unprofitable customers kept for years, discounts nobody tracks | Margin by customer, product and location, reviewed quarterly |
| 3. The bank or an investor asks for a projection and you have none | A smaller line, a higher rate, a slower yes | A three year model and a lender pack the CFO owns |
| 4. The close takes more than 15 days and the report is forty pages | Decisions made on stale numbers, meetings that argue about data | Ten day close, one page scorecard, three decisions per month |
| 5. Growth has outrun the controller: new entity, location, product line or acquisition | Tax and compliance surprises, no consolidated view | Finance structure redesign, controller promoted into their real job |
| 6. A buyer, private equity or a lender appeared and the data room took a month | A lower offer, a longer earnout, a failed quality of earnings | Seller side quality of earnings and documented add backs, 12 to 24 months out |
| 7. You cannot say what the company is worth or what would make it worth more | Years of work with no value plan | A value plan with the three or four numbers that move the multiple |
The first sign is the one most owners live with longest. Cash in the bank is a lag measure, the finish line. A CFO builds the three or four lead measures that move before cash does, and gets the owner out of the habit of logging into the bank at six in the morning to see whether the company is fine. If you are the one doing that, you are the CFO, and you are not very good at it because you have four other jobs.
The second sign is the quiet one. In nearly every $10 million company I have looked at, somewhere between a fifth and a third of the customers make no money once you load the real cost of serving them. Nobody knows because nobody has done the arithmetic. A CFO does it in the first quarter and the owner has a pricing conversation he has been avoiding for five years.
Signs three and six are the same sign at different speeds. The bank asks for a projection once a year. A buyer asks for everything at once. The IBBA Market Pulse for Q2 2026 puts median multiples at 4.0 times EBITDA for deals of $2 million to $5 million and 5.8 times for $5 million to $50 million. Every dollar of add back that fails the buyer's review comes off the price at that multiple. A $300,000 add back that cannot be documented is $1.2 million of value at four times. The CFO who documents it in advance has paid for the engagement in one finding. The quality of earnings page covers what that review tests.
Sign seven is the one owners do not see as a finance problem. They see it as a someday problem. It is a CFO problem, because the numbers that move a company's value are finance numbers: margin, recurring revenue share, customer concentration, working capital, and whether the earnings survive diligence.
What a full time CFO costs
The US Bureau of Labor Statistics does not track CFOs separately. The nearest category is financial managers, where median pay was $166,570 a year in May 2025 across 879,700 jobs, with employment projected to grow 10 percent from 2025 to 2035. The BLS lists a bachelor's degree and five or more years of experience as the typical entry requirement.
Read that median carefully. It covers every financial manager in the country, including controllers and branch finance managers at large companies. A true CFO who has run finance for a $20 million company, dealt with a bank and a buyer, and can sit on your leadership team as a peer is a different hire, and the fully loaded cost with bonus, benefits and payroll taxes sits well above the median. Then add the risk of a wrong hire at that level, which costs a year.
That is why the arithmetic for most owner led companies does not land on a full time hire first. It lands on buying the decisions you need, for the days you need them.
The three ways to fill the seat
Full time CFO. Right when the finance workload fills a week on its own: multiple entities, a live transaction, institutional investors, or a company past $50 million where the CFO is running a team of ten. Wrong when the real need is six good decisions a quarter and someone to pick up the phone when the bank calls. A full time CFO in a company with two days of CFO work a week becomes an expensive controller inside six months.
Fractional CFO. One to three days a week inside the company on a monthly retainer, owning cash, margin, the close, the lender relationship and the plan. This is the shape that fits most companies between $5 million and $50 million, and it has two advantages a full time hire cannot match. It fills the seat in weeks, and it tells you what the full time job would be before you write the job description. The fractional CFO page covers what the seat delivers and the switch points to full time. Market rates from other providers are in the fractional executive cost table, dated and sourced.
Advisory CFO. One to two hours a week with a sitting finance leader, or a fixed price two to three hour working session on one problem: the pricing model, the lender pack, the first cash forecast. Right when the controller is strong and the owner needs a thinking partner for the decisions, not another pair of hands. The CFO advisory page covers how that works, and fractional vs interim vs advisory lays the three shapes side by side.
My honest rule for choosing. Count the hours of CFO work the company has each week, not the hours of finance work. If it is under five, advisory. Five to twenty, fractional. More than twenty for two quarters running, start the full time search, and let the fractional CFO help you run it.
Where AI fits in the finance seat
AI has made the controller's job faster and the CFO's job more valuable. Reconciliation, categorisation, variance commentary and the first draft of a forecast are now hours of work instead of days. A finance leader I worked with found and fixed a floating point rounding error in a model using an AI assistant, on his own, in an afternoon. Line up a thousand finance leaders and most would not have let themselves try. That is a permission gap, not a skill gap.
What AI does not do is the CFO's actual job. It will not tell the owner that the biggest customer is unprofitable and the contract should be repriced or ended. It will not sit across from a lender and be believed. It will not notice that the owner is avoiding a conversation. Those are the judgment calls I put on the list of what not to delegate to AI, and they are the reason the CFO seat exists.
Where it changes the hiring question is this: a fractional CFO with a good AI toolset covers more of the company in two days than a full time CFO covered in five a few years ago. The AI for EOS companies page covers how companies running on an operating system are putting that to work in the weekly scorecard and the forecast. Sergey on our bench builds the data and reporting side when the stack is not ready for it.
What a CFO does not fix
A CFO does not fix a company that will not hear the numbers. I have watched a good finance leader build the margin analysis, name the unprofitable customers, and watch the owner keep every one of them because of a relationship. The CFO left inside a year. The problem was never the seat. It was the owner's willingness to act on what the seat produced.
A CFO does not fix a leadership team that cannot run a meeting. If the weekly operating meeting is the owner talking and five people nodding, the CFO's one page scorecard becomes one more thing the owner talks about. The numbers need a room that can argue with them, and building that room is leadership team development, not finance.
And a CFO does not fix the owner's dependency on himself. If every major customer, the bank and the landlord deal only with the owner, a CFO can measure that risk precisely and price what it costs at a sale. Removing it is the owner's job, with a second in command, and it takes 12 to 24 months. The exit readiness read looks at finance, operations and leadership together for that reason. A clean set of numbers on a company that cannot run without its owner is still a company that cannot run without its owner.
Who is on the bench for this
Steve Shaw is the fractional CFO on the Leaders ADAPT bench. Steve has twenty years in finance leadership and works with owners on the four things this page keeps coming back to: clean books, cash discipline, financial positioning before a sale, and the numbers conversation with buyers and lenders. He is the one who turns the forty page report into one page and the gut pricing into a margin table in the first quarter.
Thor Legvold is the leadership psychologist on the bench. When the sign behind the signs is that the owner cannot let go of the numbers, Thor's work on executive fit and the second in command is what makes the CFO seat stick.
Sergey Dmitriev is the fractional CTO and AI systems partner. When the reporting stack cannot produce a ten day close or a weekly forecast without heroics, Sergey fixes the data and the tooling and trains the team with me.
Every engagement runs through Leaders ADAPT. I take the first call, match the partner, and stay in the room.
When to hire a CFO FAQ
When does a company need a CFO?
When the decisions in front of the owner have moved past the controller's job. The usual markers are revenue between $5 million and $50 million, a bank or investor asking questions the owner cannot answer, more than one location or entity, a cash position the owner checks by logging into the bank, or a sale planned inside two years. Any two of those and the seat is open.
What is the difference between a controller and a CFO?
A controller records what happened: the ledger, the close, payroll, tax filings and compliance. A CFO decides what happens next: cash forecasting, pricing and margin, financing, the plan and the board conversation. A controller is paid to be accurate. A CFO is paid to be right about the future. Most owner led companies have a good controller and an empty CFO seat the owner fills at night.
Do I need a CFO for a small business?
Not full time, in most cases. Under about $5 million of revenue a strong controller plus an advisory CFO a few hours a week usually covers the decisions. From $5 million to $50 million the questions get bigger and more frequent, and a fractional CFO one to three days a week fits. Full time becomes worth it when the finance workload fills a week on its own.
How much does a CFO cost?
The US Bureau of Labor Statistics puts median pay for financial managers at $166,570 a year in May 2025, and a true CFO in a $20 million company sits well above the median once bonus, benefits and payroll taxes are added. A fractional CFO costs a fraction of that because you buy days, not a year. Leaders ADAPT quotes its fractional and advisory CFO work as a fixed fee after the first call.
What are the signs you need a CFO rather than a better bookkeeper?
The books are accurate but late, and nobody translates them into a decision. The owner builds the cash forecast personally. Pricing is set by gut. The bank asks for a projection and the company has none. A buyer or investor appears and the data room takes a month. Those are decision gaps, not recording gaps. A bookkeeper fixes the ledger. A CFO fixes the decisions.
Should I hire a fractional CFO or a full time CFO first?
Fractional first, in almost every owner led company between $5 million and $50 million. It fills the seat in weeks, costs a fraction of a full time hire, and tells you what the full time job would actually be before you write the job description. Switch to full time when the fractional CFO is at three days a week and still turning work away, or when a transaction is live.
Can a CFO help me sell my company?
Yes, and the work starts 12 to 24 months before a buyer appears. A CFO documents the add backs, closes the books inside ten days, builds the 13 week cash forecast and runs a quality of earnings review from the seller's side before the buyer runs one. Add backs that fail that review come off the price at the full multiple, so the CFO often pays for the engagement in the first finding.
Start with the decision you cannot make
The seven signs come down to one test. Name the last three decisions you made with a number in front of you that someone else produced, explained and recommended. If you cannot name three from the last quarter, the CFO seat in your company is empty, whatever your controller's title says.
Tomorrow morning, write down the one financial decision you have been putting off because you do not trust your view of the numbers. Pricing, the line of credit, a hire, a location. One decision. That is what the first CFO engagement works on, and how long it takes to answer tells you which of the three shapes you need.
What the first conversation looks like
It is a 30 minute call with me, not a sales rep. You bring the decision you have been putting off and the last three months of reports in whatever state they are in. I tell you whether the gap is in the books, in the decisions or in you, and whether the right first move is a fractional CFO from the bench for one to three days a week, the lighter CFO advisory shape, or a stronger controller and no CFO at all. If the honest answer is the controller, I say so.
If we go ahead, Steve joins the second call, we agree the scope in days per week and the three outcomes we write down, and the fee is fixed and quoted after that call. I stay in the room for the length of the engagement.
Your step for tomorrow morning: take the free assessments so I can see how the company depends on you before we talk about the numbers. Then book a fit conversation. Hiring a CFO is the easy part. Knowing which shape you need is what the call is for.
