Outsourced CFO arrangements fail in a fairly narrow set of ways, and one of them accounts for most of the disappointments I hear about.
It is not that the CFO was the wrong person. Usually they were impressive in the pitch and are impressive in the room. It is that six months in, the business is paying senior day rates and receiving something that looks a lot like bookkeeping supervision — and nobody can quite explain how it got there.
What happened is almost always the same thing. The engagement was specified at CFO level and the business had nothing underneath it. So the CFO, being conscientious, did what any competent person does when they find the foundations missing: they went and built them. Which is reasonable, and expensive, and not what either party intended.
Two jobs that get conflated
The distinction is worth stating plainly because the titles blur in smaller businesses.
A CFO works on the future and the outside: where the business is going, how it gets funded, what the investor or lender needs to see, which of the three options on the table is actually the right one, how the commercial model holds up under pressure. The output is decisions.
A Financial Controller works on the present and the inside: whether the numbers are right, whether the close happens on time, whether the controls hold, whether what the system says matches what is actually there. The output is reliability.
These are different jobs requiring different temperaments, and the second is a precondition for the first. A CFO’s entire value rests on the numbers underneath them being dependable — because every recommendation they make, every forecast they build and every conversation they have with a bank or an investor is downstream of data they did not produce and are not checking.
Take away the control layer and one of two things happens. Either the CFO builds it themselves, at CFO rates, for months. Or they do not, and they advise on numbers nobody has verified — which is the more dangerous outcome of the two, and the one that surfaces at the worst possible moment, usually during diligence.
The arithmetic of the mismatch
Put numbers on it, because the numbers make the argument better than the principle does.
A fractional or outsourced CFO sits broadly in the £800 to £1,200 per day range depending on scale and sector. A fractional Financial Controller runs roughly £400 to £600. An employed controller costs £65,000 to £95,000 across the UK, £75,000 to £110,000 in London, fully loaded at around 20 per cent above that — the breakdown is in Accountancy Capital’s Financial Controller salary guide.
Now consider what the work actually divides into. In a business of this size, the control layer is perhaps eight to twelve days a month of real effort: close, reconciliations, the compliance calendar, chasing and checking, keeping the ledgers honest. The CFO layer is two to four days of judgement.
If the control work has nowhere to go, it lands on the CFO. You are then buying ten days a month at £1,000 to get four days of the thing you wanted, and paying something close to a two-and-a-half times premium on the other six. Businesses rarely see this on the invoice, because the invoice says “CFO services” and the days are real. It shows up instead as a vague sense that the engagement is not delivering what the pitch described.
How to tell it is happening to you
Four signs, and any two together are usually conclusive.
Your CFO is talking about the ledger. If a meaningful share of your monthly conversation is about reconciliations, coding, chasing debtors or why the bank does not agree, you are receiving controller output from a CFO.
The forward-looking work keeps slipping. The three-year model, the funding pack, the pricing review — always next month, because this month the close ran over. Judgement work has no deadline attached and process work does, so in any squeezed arrangement the judgement work loses. Every time.
Nobody internal can answer anything. You ask a question and the honest answer is “I’ll check with the CFO.” That means the knowledge lives outside the business, which is fine as a transitional state and dangerous as a steady one.
Year-end is still an event. A business with a working control layer treats the audit as confirmation of things already known. If it remains the annual occasion on which the books are straightened out, the control layer is not there, whatever the org chart says.
What the layer has to deliver
It is a short list, and it is unglamorous, which is why it gets skipped.
A close that lands on a predictable date, every month, without heroics. A balance sheet where every line is reconciled and somebody can say what it consists of. A compliance calendar that runs without anyone chasing it. Debtor and creditor discipline that does not depend on the founder’s memory. A management pack produced mechanically rather than assembled, so the CFO’s time goes on interpreting it rather than building it. And one person who is accountable when any of that is wrong.
That last item is the one that matters most and the one most often absent. Plenty of businesses have the tasks being done by somebody. Far fewer have a person whose problem it is when they are done badly — and the difference between those two states is roughly the entire value of a controller.
Which comes first
The sequencing question comes up constantly, and my answer disappoints people about half the time.
If you have a funding round, a sale, a refinancing or a genuine strategic fork in front of you, the CFO goes first — you need the judgement now, and you accept that some of the early engagement will be spent shoring up foundations. That is a legitimate trade and worth making consciously rather than discovering later.
In every other circumstance, the control layer goes first. A dependable close and a clean balance sheet make a subsequent CFO engagement dramatically more productive, because the CFO arrives able to do CFO work from week one instead of spending a quarter finding out what is true. The businesses that get the most out of an outsourced CFO are almost always the ones that sorted the control layer before they engaged one.
And there is a third case worth naming honestly: a number of businesses that come to us wanting a CFO do not need one yet. They need reliable numbers and someone to challenge them — which is a controller, possibly with a strong Management Accountant underneath, and a CFO twelve to eighteen months later when there is a decision big enough to justify the seat.
Four ways to fill it
Roughly in ascending order of cost and permanence.
An outsourced controller function. A provider runs the process layer — transactions, reconciliations, close, compliance — with qualified oversight. Strongest on reliability, cheapest for the coverage, and the model that pairs most naturally with an outsourced CFO because the two layers stay cleanly separated. Accountancy Capital’s page on outsourced Financial Controller services sets out how these are structured.
A fractional controller. One to two days a week from an experienced FC operating inside your business rather than as a supplier. Better where the role needs presence, judgement and a relationship with the operational team; more expensive per day than a managed function.
An interim. If the real problem is that the function needs rebuilding rather than running, buy the rebuild first. An interim Financial Controller for three to six months, then a lighter steady-state arrangement afterwards. Outsourcing a mess simply produces a well-documented mess.
A permanent hire. Once the volume of genuine judgement work has outgrown what a part-time arrangement absorbs, and the knowledge needs to live inside the business. The signals that this point has arrived, and how the search tends to go wrong, are covered on the Financial Controller recruitment side.
The short version
An outsourced CFO is a judgement layer, and a judgement layer resting on nothing will either sink into the foundations or produce confident advice on unverified numbers. Neither is what you are paying for.
If your CFO engagement is not delivering what the pitch described, the question to ask is not whether you have the right CFO. It is what they have been spending their days on — and whether the thing missing underneath them would cost rather less to fix than the premium you are currently paying to have it done at the wrong rate.
The diagnostic takes an afternoon. Ask your CFO to split the last three months of their time into two columns: work that required knowing where the business is going, and work that required knowing whether a number was right. If the second column is the larger one, you do not have a CFO problem. You have a gap underneath them, and it is a cheaper gap to close than most people assume — a control layer at £400 to £600 a day, or a managed function at considerably less, against CFO time currently being spent at double or more. Accountancy Capital’s guide to outsourced Financial Controller services covers how the arrangements are priced and scoped, and where the boundary should sit between the two layers so that the split stays clean once it is fixed.
Adrian Lawrence FCA is the founder of Accountancy Capital, which recruits qualified finance professionals at £50,000 and above across the UK. He is a Chartered Accountant, holds an ICAEW practising certificate in his own name, and was previously Finance Director of a listed company. View Adrian’s ICAEW profile.
