The five days that decide the account.
Every finance client judges an outsourcing relationship on close. It is the most compressed, most visible and most overtime-heavy week of the month — and it is where a virtual desk changes the economics most sharply.
Working day 0 — the night the month ends
While the client's office is closed, the desk runs the cut-off: transactions swept, subledgers agreed to control accounts, unposted items listed, and a first pass at the accruals register from open purchase orders and recurring costs. None of it is posted. All of it is waiting when the client's controller opens their laptop.
Working day 1 — the position is visible
Bank, control account and intercompany reconciliations prepared with every break itemised and a proposed treatment attached to each. Accrual and prepayment schedules drafted from the register. Depreciation run calculated. Your reviewer works the queue; the controller attests.
Working day 2 — the awkward items
What is left is the judgement: a break that could be a timing difference or an error, an accrual with no clean basis, an intercompany balance the other side disagrees with. The desk has already gathered the evidence for each. It does not decide any of them, and it does not pretend the evidence is a conclusion.
Working day 3 — the pack
Management pack prepared: P&L and balance sheet against budget and prior period, variance analysis with the drivers identified, cash position, aged debt and aged creditors. Every figure traceable to a source document in one click, because the pack is generated from the ledger rather than retyped into a spreadsheet.
Working day 4–5 — review and the questions
The controller asks the questions a management pack always provokes. The desk answers from the underlying records within the hour rather than the day, because it does not have to go and find anything. This is the part clients notice most, and the part that renews contracts.
Working day 6 onwards — the register
Everything unresolved at close carries forward on a register with an owner and a date, not in someone's memory. The following month's close starts from that register rather than from scratch.
Why this matters to your P&L, not just the client's
On a seat model, close week is the week your margin disappears: overtime across the processing team, a senior reviewing late, and weekend cover if working day three lands badly. You absorb it because the client judges you on close and there is no way to bill for it.
A virtual desk has no overtime rate. Working day one costs what working day fifteen costs. The unit card pays per schedule prepared, so a heavy close pays more rather than costing more — which is the exact inversion of the model you run today.
What close still needs a person for.
Everything judgemental, and that is not a small residue. Whether a provision is required. Whether a break is a timing difference or an error someone needs to own. Whether an accrual basis is still reasonable given something the business did in week three that never reached the ledger.
The desk brings each of those to your reviewer with the evidence assembled and the question stated plainly. It does not resolve them, and it does not present a guess as a schedule. On a first close with a new client, expect a lot of these — the desk has no history to draw on yet, and it will say so rather than invent a treatment.
By the third close the volume of open questions typically falls sharply, because precedent decisions from the first two are indexed and retrieved. That improvement curve is real, and it is also the reason the first month feels harder than the brochure suggests.
Ask to see a close pack on the walkthrough.
Generated from the ledger, with every figure traceable to a source document. It is the single most convincing thing on the desk.