Reading this as a fractional CFO? Your clients face this exact choice, and you are usually the person they ask. This is the honest comparison, including where a hire genuinely wins.
When a mid-market CEO decides the company has outgrown its numbers, the reflex is a job ad: hire a controller. Sometimes that’s right. But the salary in the ad is the floor, not the cost — and the time cost is worse than the money. Months to recruit. Months more before the hire knows your systems. One person holding the whole function. And a reporting stack that still has to be built around them.
Here’s the full math — rough numbers, honestly drawn — and the alternative: an external controller on a governed platform, from €3,500 a month. Including the cases where hiring really is the better call.
What the hire actually costs
Start with the number in the ad. Then keep adding.
The employer pays more than the payslip. Social contributions and levies, the bonus a senior finance hire expects, benefits, equipment, software seats. Every finance director knows this math — and forgets it the moment the hire is someone for finance. All-in, the employer cost lands well above gross salary. Before the person has produced a single report.
Recruiting costs money and time. A good controller is employed and not looking. Finding one means a recruiter’s fee or months of your own effort, and from decision to signed contract is realistically three to six months for a senior profile — the same range we put on the comparison page .
Ramp-up is a hidden quarter, or two. The contract’s signed; the value isn’t. A new controller has to learn your ERP setup, your chart of accounts, your entities, your exceptions, and the story behind every adjustment. Until then you pay full cost for partial output. For a group with several entities and systems, that’s months, not weeks.
The stack still has to exist. This is the part the ad never mentions. A controller without reporting infrastructure ends up building spreadsheets — so you’ve hired an expensive person to do data-plumbing badly, or you fund the in-house build as a second project. That build is option one on the decision page , with its own three-year cost. The hire and the stack are two costs. The ad prices one.
Add it up: the real first-year cost of a productive controller is the employer cost, plus recruiting, plus the ramp-up months, plus the stack. A multiple of the ad’s number — and slower than the calendar promised.
The capacity nobody prices
Say the hire goes perfectly. There’s still a structural problem: controlling work is cyclical, and employment isn’t.
The month has a rhythm — close, reporting, review, planning, board prep. It peaks hard, then goes quiet. A full-time controller is 100% of one person every week, whether the cycle needs it or not. In practice one of two things happens: the quiet weeks fill with work below the controller’s level, or the role grows until you need a second hire.
Neither is a disaster. But both mean you’re buying capacity the only way employment sells it — one whole person at a time.
One head, again
The other problem is the one that haunts in-house BI builds: everything sits in one place. One controller means the definitions, the models, the adjustments, and the board-pack logic all live in one head. Holidays are coverage gaps. A resignation is a crisis — and it always seems to land in the worst week of the quarter. We covered this in key-person risk in finance ; a solo controller is the classic case.
You can de-risk a hire with documentation and process. Most mid-market companies never get to it — because the person who’d write the documentation is the person who’s already slammed.
The external controller model
The alternative isn’t “outsource finance.” Accounting close, statutory filings, tax — those stay exactly where they are. The external model covers the analytical and controlling layer: the monthly cycle, planning and projections, oversight of the numbers, and keeping the AI and BI layer tuned.
The forward-deployed controller works like this:
A named person, not a pool. One experienced controller who knows your business, your entities, your history. Remote by default, on-site when it matters — a board meeting, a planning round, an integration.
A set share of the week, sized to the cycle. Not 100% of one person — the share the work actually needs. Heavier around close and planning, lighter in between. Capacity that matches the cycle instead of averaging over it.
The platform is included. The controller works on the governed layer — one definition per metric, reconciled to your ledger daily , with the report library on top. The stack an in-house hire would still have to build is already there, proven across 70+ companies in 11 countries. This is the Managed tier: controller and platform together, from €3,500 a month .
A team behind the person. The named controller is the face; behind them sit the platform team, the documentation, and the monitoring. If they’re on holiday, the numbers still reconcile. The knowledge lives in the platform, not in one head — and that’s the real answer to the key-person problem, not a patch on it.
And onboarding runs the other way. The platform already holds your context — definitions, structures, entity logic — and stands it up in weeks, because the layer is built and the connectors exist for 12+ ERPs. A new hire starts from zero on your systems. The external controller starts from a working model of them.
When hiring genuinely wins
The external model isn’t always the answer. Three cases where hiring is the better call:
You need full-time capacity — really. If the work genuinely fills a week, every week — usually a larger group, heavy transaction volume, constant M&A — buy the whole person. A set share of a week is the wrong shape for a full-week job.
The role is culture-critical. Some companies need the controller in the building, in the leadership room, picking up context no remote cadence delivers. If it’s a cultural anchor and not just an analytical role, hire it.
Finance leadership is the next hire anyway. If a CFO or head of finance is coming within a year, hiring a controller now can be the first step. Then hire — and we’ll happily be the platform under them. A controller on a governed layer is a genuinely strong setup; the two work together, not against each other.
If none of the three fits — and for most companies moving from accounting to FP&A , none does yet — the honest comparison is simple: a multiple of the cost, months of delay, and one head of risk, against a named person on a proven platform in weeks. The full three-option picture is on the decision page ; the tiers are on the pricing page .
Frequently Asked Questions
What does a controller hire really cost compared to the salary in the ad? Roughly: the all-in employer cost already sits well above gross salary once you count levies, bonus, and benefits — then add recruiting, the ramp-up months, and the reporting stack the controller still needs. The first productive year costs a multiple of the ad’s number. The exact figure depends on your market and the profile, which is why we keep it directional.
Is the external controller just outsourcing under a new name? No — nothing statutory moves. Accounting close, filings, and tax stay where they are. The external controller covers the analytical layer: the monthly cycle, planning and projections, oversight, and keeping the AI and BI layer tuned — on a platform included in the price.
Who actually does the work — a person or a platform? Both, split by what each is good at. A named, experienced controller — not a rotating pool — remote by default, on-site when needed, with a team and a documented platform behind them. The platform runs the pipelines, reconciliation, and reporting; the controller does the judgment. From €3,500 a month for both together.