---
title: "How we closed 48 fractional CFO retainers worth $3,452,880 for an outsourced finance firm in six months. | Stone Haven Capital Group"
description: "An average retainer of $71,935 a year, off 628 interested founders and a reply rate we are going to print rather than explain away."
canonical: "https://stonehaven.capital/showcase/fractional-cfo-outsourced-accounting"
last-updated: "2026-08-22"
---

> An average retainer of $71,935 a year, off 628 interested founders and a reply rate we are going to print rather than explain away.

Stone Haven Capital Group

Schedule a Call

Case study · Fractional CFO & Outsourced Accounting

# How we closed 48 fractional CFO retainers worth $3,452,880 for an outsourced finance firm in six months.

An average retainer of $71,935 a year, off 628 interested founders and a reply rate we are going to print rather than explain away.

Six months. A fractional CFO and outsourced accounting firm selling part-time CFO time alongside bookkeeping and controller support on a monthly retainer. No new partners, no referral push, no events. The list ran to seven figures and the sequence ran to two emails, and both of those facts show up in the numbers below.

- $3,452,880 generated in 6 months, from 48 monthly retainers at an average of $71,935.

- 1,277,008 emails to 1,022,935 companies, producing 3,203 replies at 0.3%.

- 628 interested conversations became 194 held meetings and 48 signed retainers.

- What did most of the work: we kept every cold send away from the domain the client's finance work leaves from.

- What went wrong: we aimed the sequence at the CEO when the first real conversation belonged to somebody else.

01

## Did they get a good result?

They did, and the export it came from has not been tidied. Before the programme, new retainers arrived through the client's own network and the occasional accountant referral, worth $1,294,830 across the prior six months. The same six months with outbound running produced 48 signed retainers worth $3,452,880 in annual fees, which is 2.7 times the baseline, with 194 qualified conversations worth $8,516,600 still live behind them.

What happenedThe number

Retainers signed in six months48, worth $3,452,880 in annual fees

Average annual retainer$71,935

Qualified conversations behind them194, worth $8,516,600

Interested founders logged628, worth $17,207,200

Emails sent1,277,008

Companies contacted1,022,935

Replies3,203, a 0.3% reply rate

Positive replies628, 19.6% of all replies

Leads that finished the full sequence254,073, 24.8% of those contacted

Bounce rate2.8% on 35,756 bounces

Six months of fractional CFO outbound, exactly as the platform recorded it.

02

## Did you get them in front of the right people?

A retainer for part-time CFO time is signed by whoever owns the number at the top of the company. There is no procurement route into it and no committee to work through, so the only useful test is whether the conversations happened with founders and CEOs who could say yes on their own.

628 interested founders were logged against $17,207,200 of possible annual fees. 194 of those held up through a qualification call, carrying $8,516,600, which is 30.9% of the interested set. 48 signed a retainer, worth $3,452,880 a year at an average of $71,935 each.

The shape at the back end is where targeting shows itself. 3,203 replies produced 628 interested ones at 19.6%, and 24.7% of the qualified conversations turned into a signed retainer inside the window. A list aimed one level below the decision holds up for a reply or two and then goes quiet, because the person answering has to go and ask somebody. That did not happen here at any volume worth reporting.

A morning of interested replies landing against the fractional CFO offer.

The retainer pipeline the client's partners worked from each week.

03

## What did not work, and what we did about it

Three things went wrong, and they sit here at the front for a reason particular to this programme. Everything after this section is an argument about conversion, and conversion arguments are the easiest place in a case study to hide a mistake. Read the corrections first and you can judge the rest of the document knowing what we already admitted to.

### We aimed the sequence at the CEO when the first real conversation belonged to somebody else.

The seniority filter was set to founder, CEO and COO on the theory that the person who signs should be the person who reads. What actually happened is that a large share of the interested replies arrived from an office manager, a head of operations or whoever runs the numbers day to day, because the CEO had forwarded the email to them within a minute of reading it. The forwarded copy is the one that got answered, and it was answering a message written for somebody else.

What we changed: the operations seat went into the list as a first contact rather than as an accident of forwarding, and the follow-up was rewritten so that a person who does not own the budget can still move the conversation forward instead of stalling on it. Interested replies from that seat stopped dying at the handoff.

A forwarded thread, answered by the person the CEO passed it to.

### Sending went out on a domain that sat too close to the client's own brand.

The first tranche of sending domains was registered on near-identical spellings of the client's trading name. It felt like the careful choice at the time, since a founder checking the sender would land somewhere recognisable. The risk runs the other way. Reputation damage on a lookalike domain attaches itself to the brand it imitates, and a filtering system that learns to distrust one string is not careful about the character that differs.

What we changed: the lookalike domains were retired and the estate was rebuilt on names that stand on their own, with the client's brand appearing in the signature and the body where a human reads it rather than in the sending address where a machine classifies it. The corporate domain went back to carrying nothing but client mail.

### We launched two campaigns in the same week against segments that overlapped.

One campaign was built around agencies and the other around ecommerce, and a meaningful share of the list qualified for both. Companies received messages from two campaigns inside the same fortnight, which annoyed some of them and, worse, made both sets of results unreadable. Neither campaign could be credited with its own reply rate because neither had a clean audience underneath it.

What we changed: segments are now deduplicated against every live campaign before a launch, and launches are staggered so that each one has its own stretch of the calendar to be measured in. A result you cannot attribute is not a result.

A programme running at this width will produce a list like the one above, and any document without one has had it taken out before printing. These three are the ones most likely to appear on your engagement too.

04

## What worked, and why

### We kept every cold send away from the domain the client's finance work leaves from.

An outsourced finance firm sends month-end packs, payroll approvals and bank correspondence from its own domain every working day. A deliverability problem there is a client-service problem within hours, so none of the cold volume touched it.

We stood up 323 sending domains carrying 968 mailboxes, each one authenticated on SPF, DKIM, DMARC and MX before a single message left it, all 968 in continuous warmup and capped at 20 sends per mailbox per day. Across six months the estate carried 1,277,008 sends, finished at a 2.8% bounce rate on 35,756 bounces, and held 90.0% inbox placement with zero accounts in an error state and 16 alerts raised and cleared.

The estate that carried a million contacts, none of it on the firm's working domain.

What that buys: the domain every client relationship runs through was never exposed, and the infrastructure that took all of the risk was disposable by design.

We built the list around companies big enough to need a CFO and too small to have hired one.

Below roughly $5M of revenue a founder is still able to hold the whole picture in their head, and the retainer is hard to justify. Above roughly $50M there is usually a finance director already in the building. Between those two numbers sits a founder, CEO or COO who is making capital decisions off a bookkeeper's output. The broad pull came to 2,854,702 records on title, revenue band and geography. Enforcing the revenue band and the contact roles took it to 1,189,459, and 1,022,935 of those were actually contacted across the six months, weighted toward agencies, ecommerce and software.

Those three sectors were weighted in for a reason that has nothing to do with sophistication. All three have lumpy revenue, inventory or media spend running through the P&L, and a founder who has been surprised by cash at least once. Nothing else was filtered. No intent signals, no technographics, no funding data, because at a million contacts there is no research step that survives contact with the volume.

What that buys: revenue band tells you whether the fee is affordable and the seniority filter tells you whether anyone has to be persuaded upward. Together they are most of what a list at this width can carry, and the rest of the work moves into the offer.

We put the whole offer in the first email and asked permission rather than for a call.

The opener starts in the first sentence with no greeting at all. The second line names the kind of company the reader runs, which is the only merged detail in the message, and the ask is permission to send something across rather than a request for time. It is signed with a first name and nothing else. Read end to end it takes about four seconds, which is roughly the attention a founder gives an unknown sender at half past seven in the morning.

Every message went out as plain text, spun line by line, so no two emails left the estate identical and nothing in the body requested a remote asset.

The two-step sequence and the opener that carried it.

The opener that carried both touches, with no researched line in it.

Why that holds: a founder reading a cold email is deciding what agreeing will cost them. Asking for a calendar slot prices that at half an hour of a week they do not have. Asking whether you may send something over prices it at nothing, and it lets the interested reader raise their hand months before they are ready to buy.

We ran two touches, one wait, and routed the interested replies to a partner the same day.

Two steps with a single five-day gap between them, seven variations on the opener and two on the follow-up, all running against each other from launch. Nothing after the second email. A founder who has passed twice inside a fortnight is somebody to approach again in a later window, and against a pool this size there is always another hundred thousand companies nobody has written to once.

Why that holds: 3,203 replies over six months alongside 8,558 out of office responses at 0.8%. Two touches is enough to find the founders whose accounts are already worrying them. A third and a fourth mostly re-reach the same inboxes, and the same capacity spent on new companies finds new ones.

### We judged the openers weekly on what came back positive.

Seven opener variations against a reply rate this thin means the distance between the best and the worst version is worth hundreds of conversations across a window this long. With no research line available and no signal in the list beyond size and role, the wording was the only lever in the build, so it was reviewed every week and scored on the share of replies that were positive rather than on how many arrived. That share finished at 19.6%.

05

## The five angles we test, in every market

There is no way to know in advance which framing a market answers, and a firm that claims otherwise is describing a preference. The market decides. The only honest method is to ask it five ways at once and read what comes back. These five travel across sectors because they are built on how a decision gets made rather than on what is being sold.

06

## 1. Pain led

State the constraint they live with in the words they would use themselves, then ask whether it is true at their end. Strongest where the problem is common in the category and slightly embarrassing to raise first.

Most founders your size are making the call on hiring and spend off a set of books that arrive three weeks late. Is that roughly where you are?

07

## 2. The timing hook

Tie the message to the moment the need appears rather than to the need itself. Works wherever the buyer's problem is seasonal, event-driven or tied to a date everybody in the sector shares.

You are about eight weeks out from year end and that is usually when the gaps in the numbers show up. Worth a look before then?

08

## 3. The direct pitch

Say what you do and roughly what it costs in one sentence, then ask whether it is needed. Carries markets where the buyer is busy, commercially minded and would rather be told than warmed up.

We put a part-time CFO into companies your size for a fixed monthly fee, alongside the bookkeeping. Who does that job for you at the moment?

09

## 4. Risk reversal

Move the risk to your side of the table. Lands hardest in markets where the reader has already paid somebody for this and been disappointed.

First month is on us and there is nothing to sign beyond it. If the reporting is not better by the end of it, you have lost nothing but the introduction.

10

## 5. The benchmark

Show them where they sit against comparable companies. Almost nobody knows their own percentile and almost everybody would like to.

We pulled gross margin and runway across agencies at your revenue band. Want to see where yours lands against them?

11

## How we use them

All five launch together across a split list in the first week. Inside a month the reply data has already named the one or two this market wants and the rest come off before they consume any more of the pool. Which one wins cannot be forecast. That one or two will win can be, and running five at once converts a guess into a measurement. The same five port to LinkedIn with shorter wording and identical underlying logic.

12

## Why this works for fractional CFO specifically

Outsourced finance firms tend to grow the same way. A partner's own network produces the first dozen retainers, then accountants and lawyers refer a few more, and after that the pipeline arrives when it arrives. Nobody in the firm can tell you what the next quarter holds, which is an uncomfortable position for a business whose entire product is telling other people what their next quarter holds.

Three features of this market make outbound work unusually well:

- The buyer is one person and you can name them in advance. A retainer decision sits with a founder or a CEO at a company you can identify by revenue band and headcount before you write a word. There is nobody to route around.

- The retainer recurs and the fee justifies the volume. At $71,935 of average annual fees, a handful of signings covers the entire programme and then bills again next year without anybody sending another email.

- The need arrives on a date rather than gradually. A funding round, a bad month, an acquisition approach or a bookkeeper leaving all create the same moment. Volume is how you happen to be in the inbox during it.

13

## Why would this work for your business?

### Possibly it will not, and the table settles that faster than a conversation would.

This works ifThis does not work if

One person can sign your retainer without asking anyone elseThe decision needs a board, a committee or a procurement cycle

An engagement is worth $40,000 or more across a yearA client is worth a few thousand and never recurs

You can describe your buyer by revenue band, headcount and roleYour buyer is only identifiable after a conversation

There are hundreds of thousands of companies that fitYour entire addressable market is a few hundred names

You are content to win the ones who are ready and wait on the restYou need every interested reply to close inside the quarter

Reading down the left column means the method carries across, because nothing about this client was exceptional. A competent finance firm with good retention, real expertise and no way of reaching a founder who had not already heard their name. The process produced the result, and the process behaves the same way against your market.

14

## One more thing worth understanding

628 founders said they were interested. 48 signed. Anybody comparing those two numbers is entitled to ask what happened to the other 580, and the answer is the most useful thing in this document.

Almost none of them said no. A fractional CFO retainer is bought on trust, and a cold email cannot manufacture trust in two touches. What it can do is arrive during the week a founder is already uneasy about their numbers. When it arrives in any other week, the honest reply is that the offer makes sense and the timing does not. That reply looks like a lost deal in a spreadsheet and behaves like a warm contact in reality. A meaningful share of the 580 asked to be contacted again, and several of the 48 were people who had said the same thing in month one.

So the conversion from interested to signed here is low, and it is supposed to be. An offer that converts most of its interested replies is either priced low enough to be an impulse or is being sold to people who were already shopping. This one is neither. The programme is a search for the small number of founders standing at the exact moment where a part-time CFO stops being a nice idea and becomes this month's problem. You cannot predict who they are and you cannot schedule when they arrive. You can only be in front of enough companies that some of them are there when you knock.

That is what the volume is buying. Not persuasion, which the email is not capable of, and not qualification, which happens on a call. It buys coverage of a market wide enough that timing stops being luck and starts being arithmetic. 48 retainers and $3,452,880 came out of it, and the 580 conversations that did not close this window are the reason the second six months of a programme like this one usually beats the first.

15

## Before and after

Before After six months

New annual retainer fees won$1,294,830 $3,452,880

Retainers signed18 48

Where new clients came fromPartner network and accountant referrals 1,022,935 companies contacted directly

Qualified conversations open at period end11 194

Forecasting next quarter's new businessGuesswork based on who happened to call A funnel with 628 interested founders in it

16

## If you want to know whether your market has this in it

The first call is short. Tell us the size of company you want on retainer and what one is worth to the firm across a year of fees. We come back with how many of them are genuinely reachable, what conversation volume is realistic against that number, and whether outbound is the right instrument for this offer at all.

If we think it is not, we will say so on that call. Finding out now costs the pair of us fifteen minutes.

## Want to know whether your market has this in it?

The first conversation is short. You tell us who your buyers are and what one is worth to you. We tell you how many we can actually reach, what the meeting volume looks like, and whether outbound is the right lever for you at all.

If we think it is not, we will say so.

Book a consultation call

→

## Read another one

[Freight Brokerage & 3PL](https://stonehaven.capital/showcase/freight-brokerage-3pl/)[How we booked $5,573,986 of freight capacity for a brokerage in seven months.](https://stonehaven.capital/showcase/freight-brokerage-3pl/)[Managed IT Services Provider](https://stonehaven.capital/showcase/managed-it-services-provider/)[How we put $7,179,450 of managed IT contracts on the book for an MSP in eight months.](https://stonehaven.capital/showcase/managed-it-services-provider/)[All case studies](https://stonehaven.capital/case-studies/)