There's a scene early in Mary Poppins where the Banks children write their own job advertisement for a new nanny. Kind, clever, sweet. Someone who will listen to them. Mr. Banks tears it up and writes his own version — precise, orderly, no nonsense. Someone who will keep the household running the way he needs it to run.
He got Mary Poppins.
She was not what either of them asked for. She was what the family actually needed — which was a completely different thing, and one nobody in that house had the perspective to see clearly from the inside.
I think about this story a lot, because it maps almost exactly to what happens when a leader hires a fractional operator.
Your JD Is a Symptom Report
When leaders write a job description for a fractional hire, they're describing the problem they can see — a backlog that won't move, processes that need automating, a roadmap that needs defining and delivering, a team that needs someone to hold things together. That's the symptom. The JD is a symptom description dressed up as a specification.
I know this because I've lived it, multiple times, on multiple sides of the table.
The most dramatic example from my own career: I was brought in by an early-stage founder as a project manager. Within weeks of getting there, I found a working prototype with no product strategy underneath it — a team so deep in execution mode that nobody had ever stopped to ask the harder questions. What are we actually building? For whom? What does good look like before we build the next thing?
I couldn't help myself. I never can. I started asking why.
I wasn't a project manager for very long. The engagement became something else entirely — working through the product thesis, the 0→1 process, the decisions about what to build and what to cut. It eventually became the inspiration for the 0→1 series I wrote earlier this year, and drew heavily on the product fundamentals I'd built over years at companies like Expedia. The job I was hired to do and the job that actually needed doing were not in the same zip code.
Not Just a Tech Thing, or Even a Startup Thing — It’s a Human Thing
My sister works in commercial forestry. A few weeks ago she told me about a colleague who'd put together a scoping document for a solution the team needed. Thorough, detailed, clearly a lot of work went into it. One thing missing: any definition of the problem they were trying to solve, or the outcome they were working toward.
She'd gone straight to the solution because the solution was the part she could see. The actual problem underneath it — the decision that still needed to be made — hadn't been named yet. Maybe hadn't even been noticed.
This isn't a startup pathology or a tech industry quirk. It's just a very human default. We name what's visible. We hire for the fix we've already decided we need. We write the JD for the answer we've already landed on, before we've fully interrogated the question.
I see this pattern in almost every engagement I take on, in some shape or form.
The job description reflects what the leader thinks is broken. My first job is always to dig a little deeper and find out what's actually broken. Those two things are often not the same.
I’m Not Trying to Be Difficult — Honestly
This is the part that catches some clients off guard. They hire for execution — define the roadmap, deliver the product, automate the workflow — and what they get, at least at first, is a lot of questions. I'm genuinely not stalling or padding the engagement. I just have a deep-seated need to understand what's actually going on before I start moving things around, because in my experience, starting to execute before you've diagnosed the real problem is how you end up very efficiently solving the wrong thing.
So: what outcome are we actually trying to reach? What have you already tried? What does good look like six months from now, and how will you know when you're there? The answers to those questions almost always reframe the work — sometimes in small ways, occasionally entirely.
I'll be honest: I did have one client, early in my fractional career, who wasn't interested in that process. They had a clear vision, a specific mental model, and they wanted someone to execute it faithfully. I understand that completely — sometimes you genuinely do just need someone to build the thing you've already figured out. But as we got deeper into the work, there were signals that we might be building for a mental model the audience didn't share, and I couldn't quite let that go without raising it. They could. We eventually parted ways without any drama, but it was pretty clear we were operating with different definitions of what the engagement was actually for.
If that's what you're looking for, I'm probably not your person — and that's okay. Not every engagement is the right one.
The ones that tend to work well are the ones where the leader is genuinely curious about what they don't yet know. Where the JD was a starting point for a conversation, not the final word.
If you hire someone to execute and they start asking why — that's not a red flag. That's the job.
What Mary Poppins Actually Did
She didn't come to manage the children. She came because the family had written a job description for a problem they couldn't quite see from where they were standing. The children wrote one version. The father wrote another. Both were wrong in the same direction — they described what they thought they needed, not what was actually missing.
What was missing was somebody willing to ask the harder questions and stick around long enough to work through the answers. She fixed the thing underneath the thing, and then she left — because that's the nature of this kind of work. You're not there forever. You're there until the real problem is solved, and then the wind changes and you're on to the next one.
The leader who knows exactly what they need from a fractional hire, before the engagement starts, is rarer than you'd think. The ones who get the most out of it are usually the ones who stay genuinely curious about what that answer turns out to be.
If you've written a job description for a fractional hire recently — or you're about to — there's one question worth sitting with before you post it.
Are you describing the problem you can name, or the problem that's actually costing you?
Those might be the same thing. Probably worth finding out before someone shows up at your door with a talking umbrella and a bag that's bigger on the inside.