When to Hire a Fractional CFO: Count Backwards, Not ARR

When should a SaaS startup hire a fractional CFO?

The right answer is not a revenue threshold but a date, obtained by counting backwards from the event that will test your numbers: about 9 months before a seed closing and 12 before a Series A, and 3 months more in each case if the books are not yet in shape.

That range is not a rule of thumb. It is the sum of 3 lead times, plus a cash check, and you can settle all four against your own situation in the next 10 minutes. The 3 lead times run whether or not you are watching them. The cash check subtracts from nothing: it tells you whether the date you have in mind is reachable at all.

The reason the answer looks unfamiliar is that most guidance answers a different question. It describes a state you are already in, and you cannot act on a state. You can act on a number of months.

Why the $1M ARR trigger does not tell you when to hire

An ARR threshold measures the complexity you are already carrying, not the time you have left before the deadline, and CRV, which publishes those thresholds, calls them a rough map that non-revenue signals can pull forward.

Its guidance, published in June 2026, puts the CFO conversation in a $1M to $3M ARR window, marks $8M as the level where financial decisions typically outpace what a founder or a part-time resource can manage, and treats $15M to $20M as the outer limit past which running without one gets risky. That is a fund’s doctrine on its own portfolio rather than a measurement of the market, and it is a sensible map of when the job becomes real.

It is still the wrong instrument for timing, for one reason. A threshold is a lagging indicator. The month you cross $1M in ARR tells you nothing about how many months remain before your next financing event, and the lead time you need has already started running by then. Two companies at the same ARR, one raising in 9 months and one raising in 30, are not in the same position at all, and no revenue figure separates them.

The question is not whether you have crossed the threshold. It is whether enough months remain before the event.

How far in advance of a fundraise to hire a fractional CFO

CRV calls 3 months a reasonable minimum between engaging a fractional CFO and opening a raise, to build a defensible model, correct revenue recognition, and prepare the data room.

It is worth naming what those 3 months actually contain: clean books, revenue recognized correctly, a model that survives being pushed on, and a data room that does not generate follow-up questions for a month. That is the work of preparing to be examined, and it is genuinely 3 months of it.

But notice the assumption inside it. Those 3 months assume the books are already able to carry that work. That assumption is the part nobody prices.

The lead time nobody counts: getting the books able to carry a forecast

Those 3 months assume books that are already usable, which is not the default: across the 22 engagements with SaaS and AI companies between September 2025 and August 2026 where Offset Partners took over books previously kept by someone else, in-house or outsourced, 17 had hosting, compute, or support costs sitting in operating expenses instead of cost of revenue, and correcting the classification moved reported gross margin by 3 to 13 percentage points.

What that means in practice: a gross margin that moves 3 to 13 percentage points on a reclassification alone, 7 at the median, is not a number a forecast can sit on. If your books have been kept by someone else, in-house or by an outside provider, assume the extra lead time is there until you have checked that it is not.

How long is that extra lead time? On a 2025 engagement, bringing the monthly close from 15 to 20 business days down to around 7 took 3 monthly cycles, running in parallel with correcting revenue recognition, rebuilding deferred revenue, reconciling Stripe, and defining ARR precisely. That is 3 months, and 3 months is a material share of any raise calendar.

This is the layer that gets discovered rather than planned. Cleaning up the accounting is not CFO work, it is a prerequisite to it, which is why controller and bookkeeping work has to be in the countdown rather than treated as something the CFO will absorb along the way. A founder who budgets 3 months and finds 6 is not the victim of a slow provider. They counted one lead time and there were two.

Which case you are in comes down to two checks, and both take a minute. Does your monthly close finish inside 10 business days? Does your reported ARR reconcile to the general ledger? If either answer is no, add the second lead time.

How to count backwards to your own fractional CFO date

Count 3 lead times backwards from the month you want money in the bank: the length of the raise, the 3 months of CFO preparation, and the accounting rebuild if your close runs past 10 business days. Then run a cash check, which subtracts from nothing and tells you whether the date is reachable at all.

Define each one once, and keep the definitions fixed.

• T is the month you want cash in the bank, meaning the closing.
• Raise duration runs from the first investor meeting to cash received: 3 to 6 months at seed, 6 to 9 at Series A. These are our working assumptions from client engagements, not published benchmarks, and you should lengthen them if your last raise took longer.
• CFO lead time is 3 months minimum between engaging the CFO and opening the raise, per CRV.
• Books lead time applies only if your close runs past 10 business days or your ARR does not tie to the general ledger. 3 monthly cycles, on the engagement described above.
• Cash at opening is at least 9 months of runway remaining when you open, and closer to 12 for a Series A. Read that as a floor to open on rather than a floor to start preparing from: if you are already below it and have not opened yet, the preparation you still owe comes out of the same cash.

One trap to avoid before you start subtracting. The 6 to 9 months is a duration of work. The 12 months is a stock of cash at a point in time. They are different quantities and they do not add together. The first tells you when to start, the second tells you whether the finish is reachable.

The 4 cases, in months before T. The raise durations below use the upper bound of each range, because planning to the lower bound leaves no margin at all.

The books lead time in that table is 3 monthly cycles, which is what one 2025 engagement took, not an average. Lengthen it if your close runs worse than 15 to 20 business days, or if more than one revenue stream needs restating.

A worked example. Series A, and you want to close 12 months from now, so T is month 12.

1. The raise takes 9 months, so it has to open at month 3.
2. The CFO lead time is 3 months before opening, so the CFO starts at month 0. Today. On time, with no margin.
3. Cash check: opening at month 3 with 12 months of runway ahead of you means holding 15 months of runway today. Below that, the binding constraint is not the CFO, it is cash, and the date that has to move is the closing date. The SaaS runway calculator gives you that figure in a couple of minutes.
4. Now check the books. If your monthly close still runs past 10 business days, or your reported ARR does not reconcile to the general ledger, add the accounting rebuild: the honest start date was month minus 3. You are 3 months late, and you found out by doing arithmetic rather than by being told.

Point 4 is where most founders discover something. It is not that they were slow. It is that they were counting one lead time when there were two.

Signs you are already late to hire a fractional CFO

In 4 situations the lead time has already started rather than approaching, and each one has a named cost rather than a vague discomfort.

• Your close runs past 10 business days and you plan to raise within 12 months. The accounting lead time is already inside your window, so the CFO’s first cycles go to cleanup you are paying strategic rates for.
• An investor has asked for a data room, or a term sheet is on the table. The preparation does not shrink because your calendar did, so what compresses instead is scope: the model gets built on numbers nobody has reconciled.
• Your ARR figure comes from the CRM rather than the general ledger. The first diligence question that compares the two costs you credibility on every other number in the room, including the ones that were right.
• You have fewer than 9 months of cash and have not opened the raise yet. The constraint has changed. Another 3 months of preparation would take you to the opening with around 6 months left, and an investor who can see your cash-out date has no reason to hurry. The decision in front of you is now about cash, not about the CFO.

A 2025 engagement shows how ordinary this is. A SaaS company of about a dozen employees, billing most of its contracts annually, was planning a Series A in 9 months. The founder was sure he had a forecasting problem, so he engaged a fractional CFO directly. The books were still kept by a generalist bookkeeper, the close was taking 15 to 20 business days, and reported ARR did not reconcile to the ledger. He had not bought too early or too late by any ARR standard. He had bought the wrong layer first, and it was the accounting lead time, not the CFO lead time, that ate his calendar. It took 3 monthly cycles to bring reporting to around day 7, and the strategic work only intensified once that foundation held. The full sequence is described in bookkeeper vs controller vs CFO.

When it is too early for a fractional CFO

A company with no annual contracts, no financing event ahead of it, and a close that already finishes inside a week is buying CFO time too early, and the layer it is actually missing costs less.

Being early is a real answer and it deserves a real one. If revenue is monthly and simple, if nobody outside the company is going to examine your numbers this year, and if your reporting arrives on time and is trusted internally, then the forward-looking work has little to bite on. What usually helps at that stage is monthly accounting that stays current, and controller discipline as soon as revenue recognition gets more complicated than a single monthly plan.

The reason to say this plainly is that the countdown works in both directions. It tells you when you are late, and it also tells you when there is genuinely nothing to count yet. Two things should send you back to it: the first serious conversation with an investor, and the first month your close slips past 10 business days. Either one starts a clock that was not running before.

What to do when you are already inside the lead time

You can run the two layers in parallel instead of in sequence, cut the scope of the raise, or move the closing date. Only the first is free.

Running them in parallel is the only option that costs nothing but coordination. The accounting rebuild and the CFO work are separate jobs, but they are not strictly sequential: the model can be built against the corrected structure while the historical cleanup is still finishing, provided the same people control both and the definitions are agreed once. For a Series A with books to rebuild, it moves the start from 15 months before closing back toward 12, and no further: the CFO lead time itself does not compress. It is the reason we put both layers in one engagement rather than handing off between them.

Cutting scope means presenting less: fewer cohorts, a shorter historical restatement, a simpler model. It costs you optionality in the raise and it is sometimes the right call.

Moving the date costs runway, and it is the honest answer when the cash check in the worked example fails.

What does not work is deciding nothing. The compression happens either way. Deciding means choosing which of the three you pay.

How Offset Partners runs the CFO and controller layers on one timeline

Offset Partners holds the accounting layer and the CFO layer in a single engagement, which is what makes it possible to overlap two lead times that otherwise stack up, and the diagnostic decides up front which of the two is actually your bottleneck.

We work only with SaaS and AI companies, which is why ARR reconciled to the ledger, deferred revenue, and compute in cost of revenue are routine work here rather than something discovered and billed on your time. Fractional CFO support starts from $3,500 per month, and what moves the number is scope: how many revenue streams need schedules, whether historical cleanup is required, and whether a raise or a board cadence is running.

We will also tell you when the answer is no. If your close already finishes in a week and there is no financing event in front of you, the countdown says you have time, and we would rather say so than sell you months you do not need yet.

Sources

External figures on this page, with the date each was published.

• ARR thresholds for the CFO conversation, the $1M to $3M window, the $8M level and the $15M to $20M outer limit, and the 3 months it calls a reasonable minimum lead time before opening a raise: CRV, When to Hire a CFO, June 2026. This is an investor’s guidance on its portfolio, not a measurement, and we present it as such.
• Raise durations, 3 to 6 months at seed and 6 to 9 at Series A: our working assumptions from client engagements, not published benchmarks. No authoritative dataset exists and published estimates vary widely.
• Runway targets at opening, 9 months and 12 months for a Series A, and the 10 business day close threshold: Offset Partners guidance based on client work, not published benchmarks.
• Close times and the 3 monthly cycles: one Offset Partners engagement, 2025.
• Running the accounting and CFO layers in parallel: a choice about how we organize an engagement, not a measured result.

The classification finding, 17 of 22 engagements and a 3 to 13 percentage point gross margin correction, is Offset Partners’ own data. Population: engagements with SaaS and AI companies between September 2025 and August 2026 where we took over books previously kept by someone else, in-house or outsourced. The full figure and its method are on the SaaS runway calculator.

Related resources

• Bookkeeper vs controller vs CFO: which finance role does your SaaS startup need?
• What is a fractional CFO for a SaaS startup?
• Controller vs CFO: what does your SaaS company need?
• SaaS runway calculator
• SaaS finance readiness checklist

Related services

• Fractional CFO services
• SaaS controller services
• SaaS bookkeeping and monthly accounting
• Book a SaaS finance diagnostic