What a fractional CFO does for a funded startup, and when you actually need one

Most funded startups have reasonable records and no forecast worth showing anyone. The work of a fractional CFO starts there: a model that takes your real drivers (customers signed, average price, payroll, the suppliers you are committed to) and shows what the bank balance does over the next twelve to eighteen months under two or three honest scenarios.

That model earns its keep when you use it to make a decision. Hiring two engineers in March instead of June costs you something specific. So does moving a price, signing a longer office lease, or bringing a sales hire forward because a competitor raised. The founder makes the call. The CFO work is making sure the call is made with the number in front of you rather than a feeling about it.

The second half is the reporting rhythm. Management accounts that close within a fortnight of month end, a short board pack with the three or four measures the investors actually care about, and the same figures every time so nobody is comparing two versions of the truth across quarters. Investors stop chasing once the pack arrives on a schedule, and you stop losing a week to the chase.

A lot of the writing on this topic treats strategy and tax as separate departments. In a funded UK startup they are the same conversation, because the reliefs are the cheapest capital on the table.

SEIS and EIS advance assurance should be settled before you take the money, not after. There is a window tied to when the company started trading, and founders lose relief by raising first and asking later. We handle advance assurance in house and will speak to your angel directly, because the investor’s accountant asking a question six months on is how good rounds get slow.

Then there is R&D. If your team is building something genuinely new, qualifying expenditure comes back as a credit that improves cash. It is prepared alongside the year end here, including the CT600L and the additional information form, by the same qualified person who signs off the accounts. And if you register intellectual property, patent box can take corporation tax on profits from that IP from 25 per cent to 10 per cent, which is worth modelling long before you are profitable so the structure is already right when you are.

None of this is advisory decoration. It is cash, and it shows up in the runway line of the same model.

There is no revenue threshold that switches the need on. What we see instead is a handful of moments.

• The round has just closed and the investor update is due.
• Monthly spend has climbed to the point where a bad quarter changes the company, and nobody can say precisely when the money runs out.
• Management accounts are consistently late, so every discussion is about a position that is two months stale.
• You are about to price something, hire a team, or open a second market, and the decision is being made on instinct.
• An option scheme is coming and somebody has to handle the valuation, the paperwork and the annual return properly.

One of those on its own is usually survivable. Two or more together, and the gap stops being a reporting problem and starts being a decision problem. That is the point where senior finance input pays for itself inside the first quarter, mostly through things you do not do rather than things you do.

We will say this to founders on the first call, and it costs us work. If you are pre-revenue, pre-institutional money, spending modestly and the main thing in front of you is getting SEIS assurance in place and the company set up properly, you do not need a CFO. You need the compliance done to the letter, a basic cash forecast you maintain yourself, and the reliefs claimed on time.

Equally, bringing in a junior finance hire first is usually the wrong order. A capable generalist is excellent at keeping a system running and poor at telling a founder their pricing does not support the hiring plan, because that is a seniority question rather than a skills question. Senior judgement first, operational support underneath it once the volume justifies the cost.

The other thing worth saying plainly: a fractional arrangement ends with a short notice period. A permanent hire at that level does not, and if the fit is wrong at an early stage the cost of undoing it lands at the worst possible time. That flexibility is the real argument for the model, more than any comparison of salary numbers.

Fasih sits as a non-executive director and fractional CFO for client companies, and he runs an SEIS-backed software business of his own. He obtained advance assurance for that company before raising, so the questions an angel asks and the answers that satisfy them are familiar from both sides of the table.

In practice the compliance and the CFO work are not separated here. The person building your model is the person filing the corporation tax return, which means the forecast reflects the real tax position and the reliefs are not an afterthought bolted on in month ten. Fees are fixed and agreed in writing before anything starts, with no hourly billing and no invoices you did not expect, so the monthly cost sits in your model as a known number.

Engagement scales with what is happening. A quiet quarter between rounds needs less than the two months either side of a raise. We would rather adjust than have you paying for attention you are not using.