The operations skill is not the question. You have untangled a broken handoff between departments, stood up a hiring process from nothing, and made the trains run at a company that was barely holding together. Going fractional is a different kind of challenge: procedural, a little lonely, and front-loaded. What you actually do in the first ninety days after you stop being an employee is what decides whether the leap holds. So here is that stretch, in order.
Weeks 1 and 2: the unglamorous setup
Before a single client conversation, get the scaffolding in place. An entity, usually an LLC or an S-corp, plus an EIN and a business bank account. Errors-and-omissions and general liability insurance, because a founder's legal team will often ask for proof before they sign. A simple master services agreement you can reuse. A bookkeeping setup for your own business, which is faintly embarrassing to skip when operations is the thing you sell.
Set your rate now, in the quiet, before you are staring at a prospect. People who wait until the first call to decide what they charge tend to quote scared and anchor low. Pick a number you can say without flinching.
Month 1: turn "I run operations" into something a founder buys
"Operations" is too broad to sell. It means everything, so it sells nothing. Narrow it into a problem a specific founder recognizes. Fractional COO for seed-stage teams drowning in process debt. The operator who installs the systems layer a company needs before it can raise a Series A. RevOps and vendor cleanup for a particular stage of growth. A founder buys a named problem and a clear first move, not a title.
Write the one-pager while you are at it. What you fix, what the first thirty days look like once you are in, and what it costs. If you cannot get it onto a single page, the offer is still too vague to refer.
Months 2 and 3: landing and scoping the first engagement
The first real engagement is where the classic mistake happens. Without a full-time mandate you cannot own everything, so you have to define the slice. Two or three days a week, a written charter, a short list of what you will actually move: fix hiring and onboarding, install an operating cadence, get board reporting to a state the founder trusts. Over-scoping here is the trap. Agree to own it all and you will do a full-time job for part-time pay, resent it, and quietly conclude fractional work does not function.
Where does that first client come from? Almost always someone one degree away. A founder you supported as an employee, a peer who watched you fix something, an investor who has seen your work. Cold outreach can supplement it later. It rarely produces the first one.
Why the client problem never really goes away
Here is the part nobody warns you about at the ninety-day mark. Even once you are fully booked, engagements end by design. You were hired to fix the operational chaos, and when it is fixed, the need recedes. You are permanently replacing a client, which means sourcing new work is not a startup phase you graduate from. It is the job, running underneath the job.
That structural reality is why a referral circle is worth more than a one-time referral. It is a small group of non-competing independents, a startup lawyer, a technical recruiter, a fractional CFO, a growth marketer, all serving the same kind of company and passing work to each other as they hear about it. When a recruiter's client complains that nobody owns the process between teams, the fractional COO in the circle gets the introduction. Referna builds these groups, eight to twelve vetted peers with Gordy, our AI community manager, keeping them active, so the next engagement is surfacing while your current one is still running.