The First 90 Days of an Outsourced Finance Function
Most of the writing about outsourcing a finance function is about whether to do it. Very little is about the three months immediately after you decide, which is where the arrangements that fail almost always fail.
The pattern is consistent. A business signs with a provider in good faith, both sides are optimistic, and nothing much goes wrong in the first fortnight because the previous month’s numbers are already done. Then the first close under the new arrangement arrives, nobody is quite sure who chases the missing supplier invoices, the pack lands four days late with two balances nobody can explain, and the relationship spends the next year recovering from a start it never needed to have.
This is a guide to the transition period itself when a business moves to an outsourced financial controller arrangement: what actually happens, what the business has to do rather than the provider, and the checkpoints that tell you whether it is working while there is still time to fix it.
Why the handover is the hard part
Outsourcing financial control is not like outsourcing payroll or IT support. Those are bounded services with defined inputs and outputs. A finance function is a web of undocumented knowledge: which customer always disputes the first invoice, why the stock figure is adjusted every March, which cost centre the director’s expenses go to, what the bank wants to see and when.
Almost none of that is written down. It lives with whoever has been doing the job, and in a smaller business that is frequently one bookkeeper who is leaving, or a founder who has been doing it themselves at weekends. The provider inherits the ledger, not the context, and the first ninety days are the period in which the context either gets transferred or quietly gets lost.
The businesses that handle this well treat it as an implementation project with a named internal owner. The ones that struggle treat it as a supplier appointment and assume the provider will work it out.
Month one: the inventory
The first month should produce almost no reporting and a great deal of documentation. Resist the urge to judge the arrangement on whether a pack arrives — it should not, yet.
What needs to happen:
- A full systems and access inventory. Accounting software, banking, payroll, expenses, the CRM or job system that feeds revenue, any spreadsheets that sit between them. Access granted properly rather than by sharing a login, which is a control failure waiting to be found.
- An opening balance sheet that both sides have agreed. Every balance explained and supported. This is the single most valuable thing produced in the whole transition, because everything afterwards is measured from it — and it is the step most often skipped under time pressure.
- A documented month-end calendar. Who provides what, by which working day, to whom. Including the parts the business keeps: timesheets, stock counts, sales data, approval of accruals.
- A list of the quirks. Sit the outgoing person down and extract the undocumented knowledge before they leave. If they have already gone, budget for a worse first quarter.
Expect the opening balance sheet exercise to surface problems. Aged items that should have been written off, a VAT control account that has not been cleared in two years, a suspense balance nobody can explain. That is the arrangement working, not failing — these were already there, and the alternative is discovering them at year end.
Month two: the first real close
This is the test. The first close performed entirely under the new arrangement will be late, and it should be: plan for it rather than treating it as a breach.
What to watch is not the date but the reasons. A close that runs late because the provider was waiting on information the business did not supply is a process problem with a straightforward fix. A close that runs late because the provider could not reconcile something is a knowledge-transfer gap that needs attention now rather than next month.
Three things to insist on from the first pack, even if it is imperfect:
- A variance commentary, however short. Numbers without explanation are the thing you were trying to stop buying.
- A list of what could not be completed and why. Honesty here is the strongest early signal you will get about the provider.
- A reconciliation status summary. Which control accounts are clean, which are not, and what the plan is.
Businesses with an in-house management accountant alongside the outsourced function generally clear this hurdle a month faster, because somebody internal can answer questions in real time rather than by email. Our view on when that hybrid arrangement earns its cost is set out separately, but the transition period is the strongest argument for it.
Month three: cadence and judgement
By the third close the mechanics should be settling and the question changes. The provider can produce the numbers; the test now is whether anyone is thinking about them.
The honest distinction between a bookkeeping service and a controller function shows up here. A bookkeeping service delivers an accurate pack on time and stops. A controller function tells you the gross margin moved two points, that it is mix rather than pricing, and that one customer is responsible for most of it.
If by month three you are still receiving accurate numbers with no interpretation, you have bought the wrong thing — or you have bought the right thing and are not asking for the second half of it. Both are fixable, and the conversation is much easier at ninety days than at twelve months.
| By end of | What good looks like | Warning sign |
|---|---|---|
| Month 1 | Agreed opening balance sheet; systems access; documented calendar | “We’ll pick that up as we go” |
| Month 2 | Close completed, late but explained; reconciliation status known | Nobody can say which accounts are reconciled |
| Month 3 | Close within the agreed timetable; commentary alongside the numbers | Numbers arrive, nobody interprets them |
| Month 3 | You know who to ring and they answer | Every question goes into a shared inbox |
What the business has to keep
The most common cause of failure is not a bad provider. It is a business that outsourced the function and also outsourced its own responsibilities.
Four things cannot be handed over.
- An internal counterpart. Somebody who owns the relationship, knows what good looks like and can tell when the output is wrong. In a small business this is often the managing director, and it should be a named, deliberate decision rather than a default.
- The source data. Timesheets, stock counts, sales records, approvals. No provider can produce reliable management accounts from unreliable inputs, and “the numbers are wrong” usually resolves to “nobody did the stock count”.
- Commercial judgement. Whether to chase a customer, extend terms, take the contract. The provider informs it; the business owns it.
- Statutory accountability. Directors remain responsible for the accounts and the filings whatever the arrangement says. Outsourcing moves the work, not the duty.
The exit question, asked early
Ask at the start of the transition how the arrangement would unwind, because the answer shapes how the first ninety days should be set up.
A provider who can describe how your ledgers, workpapers, reconciliations and documented processes come back to you — held in systems you own rather than their templates — is building something you could take back in a month. A provider who is vague about it is building a dependency. The question is not adversarial; it is the clearest possible test of whether the documentation discipline you need is actually going to happen.
It is also worth checking the ordinary professional basics at the outset: whether the firm or individual holds a practising certificate with a recognised institute, who carries professional indemnity cover and at what level, and who is personally accountable if a filing goes wrong.
When outsourcing was the wrong answer
Sometimes ninety days reveals that the model does not fit, and it is better to say so than to spend two years finding out slowly.
The clearest signal is that the business keeps needing someone present. If questions arrive daily rather than monthly, if the operational team needs finance in the room, if a problem on Tuesday cannot wait until the scheduled call, then what was needed was a person rather than a service. That usually points to a fractional financial controller for one or two days a week, or a permanent hire if the volume justifies it.
The second signal is that the function needed rebuilding rather than running. An outsourced arrangement inherits whatever state the records are in and is scoped to operate a cycle, not reconstruct one. Where the opening balance sheet exercise uncovers genuine disorder, the better sequence is an interim controller for three to six months first, and the outsourced arrangement afterwards on clean foundations.
The short version
Treat the first ninety days as an implementation rather than a handover. Spend month one documenting and agreeing an opening balance sheet, expect month two’s close to be late and interrogate why, and by month three judge whether you are getting interpretation as well as accuracy.
Name an internal counterpart, keep the source data honest, and ask about the exit before you need it. The arrangements that run happily for years almost all had a deliberate, slightly tedious first quarter. The ones that collapse were the ones where everybody was too busy to do it properly.
Accountancy Capital’s guide to outsourced Financial Controller services sets out how these arrangements are structured and scoped, and where the boundaries sit against a permanent or fractional hire.
Adrian Lawrence FCA is the founder of Accountancy Capital, which recruits qualified finance professionals across the UK. He is a Chartered Accountant, holds an ICAEW practising certificate in his own name, and was previously Finance Director of a listed company. View Adrian’s ICAEW profile.
