Startups Fractional CFO Bookkeeping
Bookings, Billings, Revenue: When a SaaS Company Needs a Controller, Not Another Bookkeeper
Published November 2, 2026 by Invisible LLC Team · 8 min read
You are a founder, or you are the person at a 20-person company who ended up owning finance because nobody else would. You are not a controller, and you did not plan to become one.
Here is the version of this we hear most often: "I have the numbers. I don't trust the numbers. And the board is about to ask."
The takeaways. When bookings, billings and recognized revenue disagree, more bookkeeping hours will not fix it — the work is senior accounting judgment, not transaction volume. There are four specific triggers that mean it is time: a first audit request, a diligence list, a board metric you cannot defensibly compute, and your first material annual prepay. The right engagement is scoped, with a defined end and a written hand-off — not an open-ended retainer. And the single best thing you can do before hiring anyone is write down what "done" looks like.
Start here if you're earlier in the question. This post assumes you have already decided you need help above the bookkeeping line and you are trying to figure out whether a subscription business changes the answer. If you are still working out what the role covers at all, read our guide to outsourced controller services first — that's the canonical piece on scope, deliverables and what the function owns. We are not going to repeat it here.
Three numbers, and why they disagree
Most of the trouble in a young subscription business traces back to three words that get used interchangeably and mean entirely different things.
Bookings is what you signed. A customer commits to a twelve-month contract at $48,000 — that is a $48,000 booking, recorded the day the ink dries.
Billings is what you invoiced. If you invoiced all $48,000 up front, that is a $48,000 billing in month one. If you invoiced monthly, it is $4,000 a month.
Revenue is what you earned. The uncontested core of the rule is simple: revenue is recognized as you deliver the service. A twelve-month subscription delivered over twelve months earns revenue over twelve months, regardless of when you signed it or when the cash arrived.
So one $48,000 annual deal, signed and prepaid in September, is simultaneously a $48,000 booking, a $48,000 billing, and $4,000 of September revenue. All three are correct. None of them is wrong. And if your deck reports one number while your P&L reports another, nobody is lying — the two documents are measuring different things.
What makes this a real problem rather than a vocabulary quiz: the difference has to live somewhere on the balance sheet. The $44,000 you have collected but not yet earned is a liability. We walk through exactly how that works, journal entries and all, in what happens to your books when you sign your first annual contract. That post is the mechanics. This one is about who should own them.
The four triggers
In our experience the decision to bring in senior accounting help is almost never gradual. It arrives as a specific request from a specific person, and by then you are already behind. These are the four that actually force the issue.
1. Someone asks for an audit. A customer's procurement team, a lender, or an investor asks for audited financials. This is the one that reliably catches people, because an audit is not a thing you can do retroactively in a weekend. It requires a documented revenue policy, a defensible schedule behind every balance, and someone who can answer an auditor's questions without guessing. If this is your trigger, the useful question is not "who can get us audited" but "who can get our books to a state where an audit is possible."
2. A diligence list lands in your inbox. Twenty to forty line items, several of which you have never produced before: a revenue schedule by contract, a deferred balance rollforward, a customer-level cohort file. Diligence does not care that your bookkeeper is excellent at cash-basis bookkeeping. It cares whether the file reconciles.
3. A board member asks for a metric you cannot defensibly compute. Net revenue retention is the usual one. To compute NRR you need revenue, by customer, by period, on a consistent basis. If your revenue is recognized inconsistently — some contracts up front, some ratably, some by whenever the invoice cleared — NRR is not a number you have. It is a number you could produce, and then not be able to defend when asked how.
4. You sign your first material annual prepay. This is the quiet one, and it is the one we would watch for hardest, because it is the only trigger that arrives as good news. Cash lands, the revenue chart spikes, and your books are now accrual whether anyone told you or not. Left alone for three quarters, this becomes a restatement.
Why more bookkeeping hours do not solve it
There is a real difference between work that scales with volume and work that scales with judgment, and it is worth being precise about it rather than tribal.
Transaction work scales with volume. More invoices, more bills, more bank lines — the answer is more hours, better process, cleaner categorization. A good bookkeeper is the right answer to that problem and will be the right answer for years.
The work in front of you right now does not scale with hours. Deciding how a multi-element contract should be recognized, writing the policy that decision lives in, building the schedule that proves it, and then sitting across from an auditor and defending it — none of that gets easier if you buy twenty more hours a month of transaction processing. It needs someone who has made the judgment call before and can document why.
This is also why the framing of "we need to clean up the books" is usually the wrong frame. It implies a mess that needs tidying. More often the books are fine for what they were built to do, and the business has simply moved to a basis of accounting nobody signed off on. That is not a cleanup. It is a design change.
What a good scope looks like, and how it ends
The most important thing to get right is not who you hire. It is what you ask them for.
An open-ended monthly retainer with a vague remit is how this goes wrong. Six months later you are paying for a function you cannot describe, and the original problem — the deck and the P&L still disagree — is somehow still open. A scope with an end date forces specificity on both sides.
A defensible scope for this kind of engagement names four things:
A written revenue policy. One or two pages. How each contract type is recognized, what the triggers are, how modifications and mid-term upgrades are handled, who approves exceptions. This is the artifact everything else hangs off, and most companies at this stage simply do not have one.
A reconstructed revenue and deferred schedule. Every live contract, with the recognition pattern applied consistently, tying to the general ledger. Not a model — a schedule that reconciles.
A month-end close that produces it going forward. The schedule is worthless if it only exists once. The close has to generate it as a by-product, on a calendar, without heroics.
A hand-off plan. Who runs this in month seven? Usually your existing bookkeeper, with a documented process, a checklist, and a defined escalation path for the judgment calls. If nobody will say out loud who owns it after the engagement ends, the engagement does not end.
Ask for those four as deliverables with dates. If a prospective provider cannot tell you what the last deliverable is, you are being sold a retainer, not a project.
What to ask before you sign anything
Five questions. They are deliberately pointed, and you should ask them of us too.
- What is the last deliverable, and when? If there is no answer, there is no scope.
- Who owns this after you leave, and what are you handing them? A process document and a checklist, or a shrug.
- Are we moving off our current accounting system? For a 20-person company the answer should almost always be no. "Don't move us off QuickBooks, just make the revenue right" is a completely reasonable requirement, and a provider who insists on a platform migration as step one is solving their problem, not yours.
- Who actually signs off on the revenue policy? You want a name and a credential, not a team.
- What will you need from us, and how many hours a week is that? Every engagement like this has a founder-time cost. The good ones tell you the number up front.
Where this lands
If bookings, billings and revenue are telling three different stories in your business, that is not a sign you have been careless. It is a predictable consequence of signing annual contracts before anyone built the accounting to support them, and it happens to nearly every subscription business at roughly your stage.
What it is not is a bookkeeping backlog. It is a defined piece of senior accounting work with a beginning, a middle, and — if you scope it properly — an end.
That is the kind of engagement we take: a written policy, a schedule that reconciles, a close that reproduces it monthly, and a hand-off so the function stops depending on us. You can see the shape of the role on our controller and CFO services page, and how we work with earlier-stage companies on our startup page.
If you want to scope it before you commit to anything, tell us what's actually breaking — the audit request, the diligence list, the metric you can't defend — and we'll come back with a deliverable list and dates rather than a retainer.