Fractional CFO for startups: when to hire, the work by stage

A fractional CFO for a startup is a part-time finance leader who knows where cash is headed, can defend the runway number in a board meeting, and builds a model and a set of SaaS metrics that hold up when an investor starts asking questions. Most seed to Series B startups need that work but not a full-time CFO. This playbook is for founders of software and SaaS companies who have outgrown founder-led finance.

Bookkeeper, fractional CFO or full-time CFO

The three roles answer different questions.

• A bookkeeper answers "what happened?" Transactions recorded, accounts reconciled, books closed each month.
• A fractional CFO answers "what happens next, and what should the company do about it?" Cash forecast, operating model, SaaS metrics, board reporting and fundraising preparation, on a part-time basis.
• A full-time CFO does the same forward-looking work, plus leads a finance team every day.

A startup needs a bookkeeper first. Forward-looking finance is only as good as the books underneath it. A fractional CFO becomes worth it when the founder is spending real time in spreadsheets, a raise or board meeting is coming, or hiring decisions are being made without knowing what they do to runway. A full-time CFO usually makes sense later, when there is a finance team to run, multiple entities, or a volume of investor and lender work that needs one person's full attention.

What the work looks like by stage

The scope changes as the company grows. The goal at every stage is the same: the numbers should answer the questions the next investor will ask.

Pre-seed and seed

At this stage the company has a bank balance, a bookkeeper, and a founder who knows the business well. What is usually missing is structure.

• A clean chart of accounts and a monthly close that finishes on a predictable date.
• A simple operating model built on drivers the founder actually controls: hires, pricing, pipeline and spend.
• A 13-week cash view and a runway number under a base case and a downside case.
• A short monthly investor update with the same metrics every month.
• Clear definitions for ARR, customers and churn, written down before the first board meeting.

Series A

After a Series A, the company has a board, a plan it committed to, and investors who will compare every quarter against that plan.

• An annual budget and a monthly budget versus actual review.
• A board deck built from the same model every quarter, with commentary on what changed and why.
• An ARR waterfall, net and gross revenue retention, burn multiple and CAC payback, reported on consistent definitions.
• A headcount plan tied to the model, so each hire shows its effect on burn and runway before the offer goes out.

Series B

By Series B, the questions get sharper. Investors want to see whether growth is efficient and whether the existing customer base is compounding.

• Cohort views of retention and expansion, not just a single company-wide number.
• Scenario planning for the raise: how much capital, how long it lasts, and what the plan looks like if growth comes in slower.
• A data room and a model that tie to each other and to the books.
• Preparation for venture debt or other financing alongside the equity round, where it fits.

The SaaS metrics a fractional CFO sets up

Each metric below is shown with the standard formula. The value is less in the formula than in applying it the same way every month, so the trend means something. The full set, with what investors check for each, is in the guide to SaaS metrics investors ask for.

ARR and the ARR waterfall

Annual recurring revenue (ARR) is the annualized value of recurring subscription contracts in force at a point in time, usually excluding one-time fees, services and uncontracted usage. An ARR waterfall (or ARR bridge) explains how it moved over a period:

Beginning ARR + new + expansion - contraction - churn = ending ARR

Net new ARR is the sum of the four movements: new plus expansion, minus contraction, minus churn.

Net and gross revenue retention

Both measure recurring revenue from customers already in place at the start of the period, usually over twelve months, excluding new customers (more on NRR and GRR).

• Net revenue retention (NRR) = (beginning ARR + expansion - contraction - churn) / beginning ARR
• Gross revenue retention (GRR) = (beginning ARR - contraction - churn) / beginning ARR

GRR cannot exceed 100%. NRR can, when expansion from existing customers outweighs what was lost.

Burn rate and burn multiple

Net burn is cash out minus cash in over a period, usually monthly; gross burn is total cash out before any receipts (more on burn and runway).

The burn multiple, introduced by David Sacks of Craft Ventures, is net burn divided by net new ARR over the same period. As he describes it, the multiple asks how much a startup is burning to generate each incremental dollar of ARR: the higher the multiple, the more burn each unit of growth costs.

Runway under base and downside cases

The simple version is cash on hand divided by monthly net burn. A better version comes from a monthly forecast, because hires, lumpy collections and annual bills mean burn rarely stays flat, shown under at least a base case and a downside case.

CAC payback

CAC payback is the number of months of gross profit from new customers it takes to recover the sales and marketing spend that acquired them:

CAC payback (months) = sales and marketing spend / (new ARR × gross margin) × 12

Definitions vary on whether expansion ARR is included and whether spend is lagged by a quarter. Either choice can be defended. Changing it from one board meeting to the next cannot.

A worked example

The tables below follow one fictional Series A company through a year. Illustrative figures, not client data.

All expansion, contraction and churn in this example come from customers in the beginning ARR, so the retention figures can be read straight off the waterfall.

Illustrative figures, not client data. $K except ratios and months.

The headline here is that ARR grew 65% in a year. The waterfall tells a more useful story. Existing customers expanded by more than they contracted and churned, so NRR is above 100%. But GRR shows that 15% of beginning ARR was lost to churn and downgrades, and expansion is what covered it. If expansion slows, the base erodes. That is the question a good investor will ask first, and a founder should already have the answer.

How an investor reads it

Decipher sets up a startup's finance the way an investor or acquirer will read it. In diligence, the same few things get checked, regardless of how polished the deck is.

• Does the waterfall tie? Beginning ARR plus the movements should equal ending ARR, and ending ARR should reconcile to the contracts and the books. A gap that cannot be explained undermines every other number.
• Where does growth come from? NRR and GRR together show whether growth comes from new logos, from expansion, or from a base that is leaking. Two companies with the same ARR growth can look very different here.
• What does growth cost? The burn multiple and CAC payback put growth next to the cash spent to get it.
• Are the definitions stable? If ARR included services last quarter and excludes them this quarter, the trend is meaningless even when each figure is technically correct.
• Does the forecast hold up? Last year's plan compared with actual results says more about a team than next year's projection.
• Is runway honest? A runway number with only a base case, or one that assumes the next round closes on time, tends to get discounted.

Fundraising preparation

A fractional CFO usually starts preparing for a raise months before the first pitch meeting. The core pieces:

1. The model. A driver-based operating model with an ARR waterfall, hiring plan, and cash forecast, showing how much capital the plan needs and how long it lasts.
2. The data room. Financial statements, the ARR and customer data behind the metrics, cap table, key contracts, and a metric definitions page. Everything in it should tie to the model.
3. The board deck. The same structure every quarter: results against plan, the metrics, cash and runway, and the decisions the board needs to weigh in on.
4. Investor updates. A short, regular update with consistent metrics. Founders who send them before a raise start the process with investors who already know the trend.

The 13-week cash view

The operating model answers questions over quarters and years; the 13-week cash view answers whether the company will have the cash it needs each week for the next three months. For a startup with less than a year of runway, or one waiting on a few large customer payments, it is often the most useful single report, and the 13-week cash flow forecast guide shows how to build one.

Common mistakes

• Counting bookings or one-time fees as ARR. It inflates the number and creates a correction later, usually in diligence.
• Reporting NRR without GRR. A strong NRR can hide meaningful churn behind expansion from a few large customers.
• Starting fundraising preparation when the raise starts. The data room and clean metric history take months to build, not weeks.

What to look for in a fractional CFO for a startup

• SaaS experience. ARR, retention and burn multiple should be familiar ground, not something to learn on your engagement.
• The investor's perspective. Someone who has seen how investors and acquirers test a model will build one that survives that test.

The full checklist, including who does the work and how they fit with the bookkeeper and CPA, is in the guide to how to choose a fractional CFO. See Decipher's fractional CFO services for how engagements are scoped, or take the finance scorecard, which takes about two minutes and shows where the gaps are today.