“Fractional sales arm” is a category name, not a description of work. Ask five providers what it means and you will get five monthly numbers and a promise about pipeline.
This is what it actually looks like from the inside — the sequence, the artefacts, and what you should be able to see at each point. If a provider cannot show you this, you are not buying a system. You are buying activity.
Weeks one and two: calibration
Nothing gets sent in the first fortnight. That is deliberate, and it is the part most engagements skip.
The commercial download. Not a discovery call. Several hours with whoever knows the deals — usually the founder — going through the last dozen opportunities: which closed, which stalled, what the objection actually was, and what the buyer said in the room rather than what the CRM recorded. Most of the message comes out of this conversation, not out of the website.
Buyer definition, sharply. Not a persona document. A defined list of company types, sizes, sectors and triggers, with the reasons attached. The output has to be specific enough that a name either qualifies or does not — “mid-market B2B” is not a definition, it is a hope.
Infrastructure. Sending domains registered and configured, authentication set up properly, mailboxes warming, CRM fields mapped, tracking in place. Warm-up takes weeks and cannot be compressed, which is the main reason the first fortnight exists. Sending from cold infrastructure is how a programme damages a domain in week one and spends three months recovering.
The first message set, written and argued about. Drafted by the person who will send it, then pulled apart with you. This is where most of the disagreement in an engagement happens, and it should — it is far cheaper to argue about a claim now than to test it on four hundred buyers.
What you should see by day fourteen: a written strategy, a defined target list, live infrastructure, and message drafts you have read and challenged.
Weeks three and four: first contact
Volume starts low on purpose — a controlled ramp that protects deliverability and, more importantly, produces information before scale.
The first sends go to a deliberately small segment. Not to be cautious, but because the first hundred sends are a test instrument. Which framing gets replies. Which objection appears immediately. Whether the buyer definition survives contact.
Replies get handled the same day, by the person who wrote the message. This matters more than the message itself. A reply is a live buyer, and the difference between a same-day response from someone who knows the offer and a next-week response from a coordinator is the difference between a meeting and a dead thread.
What you should see by day thirty: real reply data, the first meetings booked or in motion, and an honest read on whether the message is landing — including the parts that are not.
The first hundred sends are not a campaign. They are an instrument, and what they measure is whether you were right.
Month two: the correction
Month one produces a hypothesis. Month two is where it gets corrected, and this is the month that separates a system from a subscription.
Everything gets re-examined against actual replies. Which segments responded and which went silent. Whether the seniority you targeted was right — it frequently is not, and the correction is rarely upward. Which of your claims buyers pushed back on. Which objection recurs often enough to belong in the opening rather than the rebuttal.
The message gets rewritten, not tweaked. Volume increases where the evidence supports it and stops where it does not. Segments that produced nothing get retired rather than retried harder.
You should expect the month-two report to contain something uncomfortable. If a provider’s second month is a clean report about steady improvement, either the first month was suspiciously perfect or nobody is looking properly.
Month three onwards: compounding
By the third month the shape is set and the work changes character. It becomes less about discovery and more about rhythm:
- Weekly: sends go out, replies are handled same-day, meetings are booked and briefed back to you, and the list is refreshed with new accounts and new triggers.
- Fortnightly: message variants tested against each other, one variable at a time.
- Monthly: the four metrics — plus, more usefully, what changed in the message and why.
- Continuously: deliverability monitoring, list hygiene, and the discipline of retiring sequences before they decay.
This is also where a well-run programme stops being purely outbound. The reply data tells you what your market actually objects to, which is the most useful marketing input a business gets all year — and it usually changes the website before it changes anything else.
A realistic expectation: the first month is calibration, the second is correction, and the third is when pipeline built on a corrected message starts landing. That is why the engagement carries a six-month minimum. A shorter engagement mostly measures the learning phase and calls it a result.
What we need from you
The engagement fails predictably when three things are missing.
Someone who can answer questions inside a day. Not a committee. One person with commercial authority who can confirm whether a claim is true and whether a segment is worth chasing.
Willingness to have the message challenged. The message that closes is rarely the message the business is attached to. If your positioning is not open to being wrong, the programme is capped at what you already know.
Taking the meetings seriously. We book conversations; you close them. A meeting rescheduled twice is a dead meeting, and a buyer who was warm three weeks ago is not warm now. This is where more programmes underperform than any messaging issue.
The bottom line
The value of a fractional arm is not that someone else sends the emails. It is that the person who set the strategy is the person handling the replies, so what the market says goes straight back into what gets sent — with no handover, no translation, and no quarter lost to a briefing document. The mechanics behind that, including how it is priced, sit on our sales outreach page.
Ask any provider for their version of this article. The ones running a system will have one. The volume model does not produce one, because there is nothing to describe past the send.