Fractional PM, agency, or full-time hire: what a startup actually needs
Fractional to find out what to build, full-time to keep building it, an agency when you already know exactly what you want. How to tell which stage you're at.
Every founder hits the same moment. The product has stopped being something you can hold in your head, and engineering is waiting on answers nobody in the room is qualified to give. You need product leadership. The question is what shape it should take, and the honest answer depends on stage, budget, and how much of the problem is still unknown. I run a product studio in Vancouver, so I have an obvious bias. I have also been the full-time hire, so here is the comparison I would give a friend.
What each option is actually selling
A full-time product manager sells continuity. They are in every meeting, and after a year they are the person who knows why every decision was made. That memory is what you are paying for, and it takes months to build.
An agency sells throughput. You describe the outcome, they staff a team, and something ships. The catch is that the people who did the thinking leave when the invoice is paid.
A fractional product manager sells judgment by the week. You get a senior person for a day or two a week who has done this before and will make the calls and write the specs, without the cost of a full seat. They are also not in the building every day, which matters more at some stages than others.
The stage test
Before product-market fit, the dominant risk is building the wrong thing. That risk is not reduced by adding engineers, and it is only partly reduced by adding a full-time PM, because a full-time PM at that stage spends most of their first quarter learning your domain. What reduces it is discovery: interviews, and requirements written as acceptance criteria someone can test. That is a bounded piece of work, and bounded work is what fractional and studio models are built for. It is also the shape of nearly everything ProductGurus does for clients: a discovery sprint, a prototype, or both, with the production build left to your own team.
After product-market fit, the dominant risk flips to execution and continuity. Now you need someone who is there when the support ticket comes in, who owns the roadmap conversation with sales, and who will still be there in eighteen months. That is a full-time hire, and no amount of fractional cleverness substitutes for it.
Where each one goes wrong
The full-time hire goes wrong when it happens too early. A founder hires a PM at seed, hands them an ambiguous product, and six months later has paid a full salary for a roadmap document and a lot of meetings. Usually the person was fine. The timing wasn't.
The agency goes wrong on ownership. Agencies are good at building what is in the brief and have no commercial reason to ask whether the brief is right. When GridSentinel discovered that its headline feature couldn't work on the line it was being pitched for, the useful response was to re-scope the product. An agency paid to deliver the original spec would have delivered the original spec.
The fractional PM goes wrong on availability and on scope creep. If the engagement quietly becomes "be in every standup," you are paying senior rates for a part-time employee, and nobody is happy. The fix is to define the deliverable up front, whether that is a discovery report or a shipped v1 with documented outcomes. Every project on our work page is structured that way, mostly because it keeps both sides honest.
Rough economics, without the sales pitch
In Vancouver, a mid-to-senior full-time product manager is a six-figure commitment once you add benefits and the months before they are productive. An agency engagement for a real MVP typically starts in the tens of thousands and climbs quickly with scope. A fractional arrangement is usually priced by the day, for a set number of days a week over a defined period, and comes out to a fraction of that salary.
None of those numbers is the expensive one. The expensive one is three engineers for six months on a feature nobody wanted, and it never shows up on a quote. Whatever model you pick, spend the first money on making sure that doesn't happen.
A simple way to decide
If you can describe your product's next twelve months with confidence, and the gap is purely capacity, use an agency or hire engineers. If you cannot describe the next three months with confidence, bring in fractional product leadership for a defined discovery-and-spec engagement, and decide about the full-time seat when you know what the job is. If you have paying customers and a roadmap that other people depend on, make the full-time hire and give them a real mandate.
And if you are somewhere in the middle, which most companies are, the sequence that tends to work is fractional first to define the thing, then a full-time PM who inherits a product with its reasoning already documented. That handover is far easier when the discovery exists on paper instead of in someone's head, which is why every ProductGurus case study reads more like a spec than a highlight reel.
Sources and further reading
- Inspired and Empowered, Marty Cagan (SVPG). The standard reference on what a product manager is actually accountable for.
- GridSentinel case study. The re-scope referenced above.