Fractional CFO Bookkeeping Startups

Outsourced Controller Services: What They Cover, and When a Business Actually Needs One

Published October 5, 2026 by Invisible LLC Team · 11 min read

You didn't start a company to own a month-end close calendar. But somewhere between "my bookkeeper handles it" and "we should probably hire a CFO," a gap opens up — and it usually announces itself the same way. The books close, eventually. The numbers are probably right. Nobody can tell you why the number moved, and nobody owns the answer.

That gap has a name, and the name is controller.

Key takeaways: An outsourced controller owns the monthly close, the accuracy of the financial statements, and the controls underneath them — the layer between transaction-level bookkeeping and CFO-level strategy. The job is defined by ownership of a deliverable calendar, not by hours. And the honest trigger test isn't revenue: it's whether a specific person can currently be held responsible for the close date, the reconciliations, and the reporting package. If the answer is "sort of, between three people," that's the gap.

What outsourced controller services actually cover

Vendor pages tend to describe this work as "oversight," which tells you nothing. Here is the scope in the terms a buyer can actually check against:

Owning the monthly close on a stated calendar. A real close has a date, a checklist, and a person whose name is on it. Books closed by the 10th business day means something. "Closed when the bookkeeping catches up" does not.
Reconciliations, all of them, every month. Bank and credit card, merchant processor and payout accounts, loans and lines of credit, payroll clearing, intercompany if you have more than one entity. Unreconciled accounts are where surprises live.
Accrual discipline. Accruals, prepaids, deferred revenue, fixed assets and depreciation schedules, and the cut-off decisions that put a transaction in the right month. This is the single biggest difference between books that are current and books that are right.
The reporting package. P&L, balance sheet, cash flow, budget-versus-actual, and — this is the part that separates a controller from a very good bookkeeper — a written explanation of the variances. The number plus the reason.
Internal controls and segregation of duties. Who can approve a bill, who can release a payment, who reconciles the account those payments came out of. In a small company these three are often one person, and that one person is usually the owner.
AR and AP oversight. Not the keystrokes — the aging, the collection cadence, the approval workflow, the vendor master file.
Payroll and benefit accruals booked correctly, especially when you've hired across state lines and the payroll provider's export doesn't map cleanly to your chart of accounts.
Audit, review, and tax-preparer support. The controller is who assembles the schedules an outside CPA or auditor asks for, and who answers their questions without derailing the month.
System and process ownership. The chart of accounts, the close checklist, the documentation that means the process survives a staffing change.

If you want the role boundaries laid out cleanly — controller versus bookkeeper versus CFO, and what each one is actually accountable for — we wrote that up separately in what a fractional controller does versus a fractional CFO. This post assumes you've already decided the controller layer is the one you're missing, and is about scope and timing.

What outsourced controller services don't cover

The boundary matters more than the scope, because most bad engagements come from a buyer assuming something was included.

It is not an audit, a review, or any form of assurance. Those are engagements performed by an independent CPA firm under professional standards. A controller prepares for the audit — assembles the schedules, answers the questions, keeps the records defensible — but does not perform one. Any firm describing both as a single engagement is describing two different things.
It is not tax return preparation. A controller keeps the records in the shape your tax preparer needs and hands them over clean. Filing the return is a separate engagement, usually with a separate firm.
It is not CFO work. Fundraising strategy, pricing architecture, scenario modeling, board narrative — that's the layer above. Plenty of firms sell both; they're still different jobs with different deliverables, and paying controller rates for CFO judgment (or the reverse) is the most common scoping mistake we see.
It is not a framework consultancy. If you need SOC 2 built from scratch, that's a specialist engagement. A controller keeps the financial controls documentation auditable on an ongoing basis once the framework exists — a real and useful contribution, but not the same purchase.
It is not a headcount replacement for a broken bookkeeping function. A controller supervising bad source data produces beautifully formatted wrong answers. The transaction layer has to work first.

Five triggers that mean you need one now

Revenue thresholds are the wrong test. Two companies at the same revenue can have wildly different finance needs depending on entity count, billing model, and how many states they employ people in. These triggers are better predictors.

1. Nobody can name the close date. Ask three people at your company when last month's books closed. If you get three answers, or one shrug, the ownership gap is already real. This is the cleanest single diagnostic in the list.

2. You've added complexity your bookkeeping wasn't built for. A second entity. A second currency. Subscription billing with deferred revenue. Inventory. Payroll in four states. Each one adds a monthly judgment call, and judgment calls are the controller's job, not the data-entry layer's.

3. Someone outside the company is about to read your financials closely. A lender, a landlord requiring financials for a lease, an acquirer, a grantmaker, or an investor running diligence. Outside readers ask questions that are trivial to answer from controlled books and nearly impossible to answer from books that were merely current. If a raise is what's coming, start with our pre-raise checklist for founder-kept books — it's the companion piece to this one and it's specific about what breaks.

4. You're making real decisions on numbers you don't fully trust. Hiring, pricing, restocking, opening a location. The tell is behavioral: you have a monthly P&L and you still make the call from your bank balance.

5. Your best finance person has become the bottleneck. In-house teams hit this constantly — a capable accountant or ops lead is underwater on the technical work (revenue recognition, deferred revenue, multi-entity consolidation) and everything else queues behind it. The fix isn't always another full-time hire; sometimes it's taking the two hardest workstreams off their plate on a defined scope with a hand-off plan.

The in-house comparison, honestly

The alternative to outsourcing isn't "nothing." It's hiring, and hiring has a real market price.

The U.S. Bureau of Labor Statistics groups controllers inside its financial managers category, and reports a median annual wage of $166,570 as of May 2025, with the lowest 10% under $94,310 and the highest 10% above $323,270 (Bureau of Labor Statistics, Occupational Outlook Handbook). BLS also projects employment in that category to grow 10% from 2025 to 2035, faster than the average across all occupations — meaning this is not a role that's getting easier to hire.

Two honest caveats about that figure, because it gets misused in both directions. It's a wage — payroll taxes, benefits, software, and recruiting sit on top of it. And it's a national median across a broad category that includes finance roles well above and below a small-company controller. Use it as an order-of-magnitude anchor for the decision, not as a quote.

The real trade-off isn't cost, though. It's fit to the shape of the work. A controller-level workload at a company doing a few million in revenue is usually not forty hours a week of controller-level work — it's ten to fifteen hours of genuinely senior judgment plus a lot of process. Hire full-time for that and you either overpay for idle senior capacity or you fill the gap with tasks the role is overqualified for and the person quietly gets bored. Outsource it and you buy the senior hours without the seat.

Where in-house wins, clearly: when the finance function needs to be physically and culturally embedded, when the volume genuinely supports a full-time senior person, or when you're building toward an in-house team and want the first hire to grow into it. Any firm that tells you outsourcing is always the answer is selling, not advising.

How to scope the engagement so it doesn't drift

Retainer creep is the most common failure mode in this category, and it's almost always a scoping failure at the start rather than bad faith later. Insist on these five things in writing.

A named close date. Not "monthly reporting." Books closed and the reporting package delivered by the Nth business day. Everything else in the engagement hangs off this.
The deliverable list, itemized. Which statements, which supporting schedules, which reconciliations, what cadence, and in what format. If budget-versus-actual with written variance commentary matters to you, it goes on the list.
A response-time expectation. Same-day or next-day on questions, in the channel you actually use. A controller you have to schedule a meeting with to ask a question is not functioning as part of your team.
The boundary, stated. What's included, what's a separate project, and what triggers a scope conversation instead of a surprise invoice. Cleanup of historical books is project work and should be priced and scoped as project work — not folded silently into month one.
A hand-off plan, even if you never use it. What documentation exists, who holds the credentials, and what a transition to an in-house hire would actually involve. A firm that can describe its own exit is a firm that documented the process.

One market data point worth knowing as a buyer: among CPA firms with client advisory practices surveyed by CPA.com and the AICPA, only 10% still used hourly billing as their primary pricing method, down from 53% in 2018 (Journal of Accountancy, December 2024). The same survey reported a 17% median revenue growth rate for these practices in 2023. Translation for you: fixed-scope, fixed-fee engagements are now the norm in this category, so an hourly-only proposal is worth asking about — sometimes it's the right structure for an undefined cleanup, and sometimes it means the scope was never pinned down.

Seven questions to ask before you sign

Who specifically does the work, and who do I talk to? Get names and the distinction between them. "A team" is not an answer.
What day do my books close, and what happens if you miss it?
Walk me through your close checklist. A real one exists and can be shown. A vague answer here is the single strongest negative signal in the whole conversation.
What does your monthly package include, and can I see a redacted example?
Who else in my industry, at roughly my size and complexity, are you doing this exact work for? Ask directly. And notice whether they answer honestly when the answer is "nobody exactly like you."
What's explicitly not included? If they can't list three things, they haven't scoped it.
What would the hand-off look like if I hire in-house in eighteen months?

Where we fit — and where we don't

We run this layer for owner-operated companies and founder-led startups: a stated close date, reconciliations that are actually done, accrual discipline, a reporting package with the variances explained in plain language, and AR/AP oversight underneath it. It ladders directly off the bookkeeping, which is why it works — the controller layer is only as good as the source data beneath it. You can see the full scope on our fractional controller and CFO services page.

Being straight about the edges: we are not an audit firm and we don't issue assurance. We're not a SOC 2 framework consultancy. And for a venture-track company past Series A with a $5M+ run rate and an in-house finance lead already in place, you've likely outgrown our altitude and we'll tell you so on the first call rather than three months in.

What we are is the senior judgment layer for companies that need it and can't yet justify the seat — the person who owns the close, explains the number, and answers on the same day you asked.

Tell us what your close looks like today and we'll tell you honestly whether you need a controller, better bookkeeping, or nothing yet. If you're a founder heading into a raise, the companion piece is our pre-raise diligence checklist; if you want the role definitions first, start with fractional controller versus fractional CFO. You can also see how we work with startups and founders or what the custom reporting layer looks like once the close underneath it is reliable.