Your banker asks a simple question at the annual review: what will your cash position be at the end of the third quarter? You have a bookkeeper who is excellent at paying bills, an outside CPA who sees the numbers once a year, and a set of financials that closed on the 24th of last month. Nobody in the building can answer the question, and the people who could answer it do not trust the numbers they would have to start from.
That moment is where the Controller versus CFO question usually gets asked, and it gets asked badly, because the two titles are treated as a ladder. They are not a ladder. They are two different jobs that happen to share a spreadsheet. This article lays out what each role owns, a symptoms table to diagnose which gap you have, and how to sequence the fix when you have both gaps at once, which is more common than anyone likes to admit.
What a Controller owns
A Controller is accountable for the accuracy and timeliness of the past. The job is the close, the reconciliations, the controls, and the process that makes the numbers trustworthy. A good Controller runs a monthly close that finishes within 10 business days, reconciles every balance sheet account, not just the bank, and produces a financial package that means the same thing this month as it did last month.
The Controller also owns the machinery underneath: chart of accounts, revenue recognition mechanics, job costing or class tracking, fixed assets, payroll liabilities, sales tax, intercompany, and the audit or review support when the CPA shows up. In a construction company, the Controller owns the WIP schedule and the monthly over and under billing entries. In a distributor, the inventory count and the cost layers. The work is unglamorous, precise, and the entire foundation for everything a CFO does. A forecast built on unreconciled books is a guess with formatting.
What a CFO owns
A CFO is accountable for the future and for the money. The job is the forecast, the capital structure, the lender and investor relationships, pricing and margin strategy, and the decisions that change the shape of the business: whether to buy the competitor, open the second location, take the bank's term sheet, or exit the unprofitable service line that everyone is emotionally attached to.
The CFO builds and owns the 13-week cash-flow forecast, the annual budget and rolling reforecast, the KPI framework that tells the owner what actually happened before the financials do, and the covenant calculation that keeps the bank calm. In a turnaround, the CFO runs the cash, negotiates with creditors, and produces the reporting a lender or a court requires. The CFO needs a reliable close to do any of this, which is why the CFO is the second hire, not the first, when the books are in doubt.
The symptoms tell you which gap you have
Owners rarely describe the problem in role terms. They describe symptoms. The table below maps the symptoms we hear most often on a first call to the role that fixes them. Read down the list and count where your marks land.
| What you are experiencing | Points to | Why |
|---|---|---|
| The close takes more than 15 days, or never truly finishes | Controller | Close discipline is a process problem, and process is the Controller's job |
| Prior months get restated after you have already reviewed them | Controller | Restatements mean reconciliations are happening late or not at all |
| The balance sheet has accounts nobody can explain | Controller | Unreconciled accounts are the classic sign of a bookkeeper working above capacity |
| The CPA does a large cleanup every year at tax time | Controller | The books are not being maintained to a standard the tax return can use |
| You cannot say what cash will be in eight weeks | CFO | A cash forecast is a finance deliverable, not an accounting one |
| The bank asked for a forecast or covenant package you could not produce | CFO | Lender reporting and covenant management are CFO responsibilities |
| You are profitable on paper and short on cash | CFO | This is a working capital and billing structure problem, which is CFO territory |
| You do not know which jobs, customers, or lines make money | Both | The Controller builds the costing; the CFO decides what to do with it |
| You are considering an acquisition, a new location, or a sale | CFO | Capital allocation and transactions are CFO work by definition |
| Growth is fast and the finance function is not keeping up | Both | Volume breaks the close first, then breaks the forecast |
| Multiple entities and the intercompany balances do not tie | Controller | Consolidation mechanics are accounting, and untied intercompany means the books are wrong |
| You are making pricing decisions from gut feel | CFO | Pricing and margin strategy require modeling, not bookkeeping |
If most of your marks land in the Controller column, do not hire a CFO yet. If they land in the CFO column and the books are actually clean, you have earned the right to skip ahead. If they land in both, keep reading, because the sequence matters more than the choice.
A decision framework in four questions
When the symptoms table is ambiguous, four questions resolve it. Answer each honestly, preferably before the next bank meeting rather than during it.
- Do you trust the balance sheet? Not the income statement. The balance sheet. If any account on it would embarrass you in a due diligence call, the answer is no, and the first need is a Controller.
- Is the question you cannot answer about the past or about the future? Past means accounting. Future means finance. Most owners have one of each, but one is usually more urgent.
- Who is asking? A CPA asking about the books wants a Controller. A lender, investor, or bonding agent asking about the outlook wants a CFO.
- What decision is pending? If the next 12 months include a capital raise, an acquisition, a major hire, or a lender renewal, you need CFO-level work whether or not you have a title for it.
A useful shorthand: the Controller makes the numbers right, and the CFO makes the numbers useful. You cannot do the second without the first, and doing the first without the second leaves you with a beautifully reconciled set of books and no idea what to do on Monday.
When you need both
Companies between $3M and $50M commonly have neither role filled properly. The bookkeeper has been promoted to Controller by title without the training, and the owner has been acting as CFO without the time. Both gaps are real, and the tempting move is to hire one person to be both. That person is rare, expensive, and, in our experience, ends up doing the accounting because the accounting is what screams loudest each day.
The better structure is a Controller who owns the close and a CFO, fractional or full-time, who owns the forecast and the outside relationships. The two roles have different rhythms. The Controller lives in the first two weeks of the month. The CFO lives in the bank meeting, the forecast update, and the quarterly decision. When one person tries to hold both rhythms, the forecast is the thing that slips, because nobody outside the building notices a missing forecast until the bank does.
Cost matters here. Our fractional Controller engagements start at $2,500 a month and fractional CFO engagements start at $3,500 a month. Together, that is often less than a single full-time hire with the title Controller and the expectations of a CFO, and it covers both jobs properly rather than one job well and one job on weekends.
The sequence: Controller first, unless
When both gaps exist, the default sequence is Controller first. The reasoning is mechanical. A CFO's first deliverable is a cash forecast, and a cash forecast starts from the accounts receivable aging, the accounts payable aging, and the bank balance. If those three do not reconcile to the general ledger, the forecast inherits the errors, and the lender who receives it will find them before you do.
There are three exceptions. First, if cash is critical right now, the forecast cannot wait for a clean close, and a CFO builds a direct cash forecast from the bank and the agings while the cleanup runs in parallel. Second, if a transaction is already underway, the buyer's diligence team will set the timeline for you. Third, if a lender has already asked for something, the answer is the thing the lender asked for, delivered on time, with a note about what is being fixed underneath. In each case, the CFO leads and the Controller work follows, but it follows immediately, not eventually.
What the titles look like in practice
A picture helps. In a $12M specialty contractor with 45 employees, a Controller runs a close that finishes by the 10th, maintains the WIP schedule monthly, reconciles retainage receivable and payable, and produces a package the bonding agent can read without calling. The CFO takes that package, updates the 13-week cash forecast every Monday, models the effect of the two large bids in the pipeline on working capital, and negotiates the line of credit increase before the bids are won rather than after the payroll is short.
In a $6M professional services firm with three partners, the Controller is often part-time, the close is simpler, and the CFO work concentrates around pricing, utilization, partner compensation, and the decision about whether to open the second office. The roles scale with the business, but the division of labor does not change: one role is accountable for the numbers being right, and the other is accountable for the numbers being used.
Where a fractional Controller or CFO fits
The fractional model exists precisely because most companies in this range need both functions and can justify neither as a full-time executive. A fractional Controller installs the close, reconciles the balance sheet, documents the process, and either trains your existing staff to run it or hands it to a full-time hire when the volume justifies one. A fractional CFO builds the forecast, owns the lender relationship, installs the KPIs, and shows up for the decisions that change the business.
The engagement usually starts with an assessment of which gap is larger, then a cleanup and stabilization phase that gets the books to a standard a forecast can be built on, then the forecast and reporting cadence, then documentation and transition. If you already know which column your symptoms land in, you already know where to start. If you do not, that is what the assessment is for.