Building a Back Office That Does Not Eat Your Margin

A perm-only agency can run its back office on a bookkeeper and a spreadsheet. A contract desk cannot, and the transition catches firms out because it happens gradually. Timesheets multiply, invoicing becomes weekly, statutory obligations accumulate and cash collection turns into a job. Firms discover their back office is the constraint when growth stops for reasons nobody can locate on the sales side.

Last reviewed August 2026. Assessment drawn from Sortinghat customer observation, stated as directional.

Why it becomes the constraint

A perm placement is billed once. A contract placement generates a timesheet, an invoice and a payment cycle every week for the life of the assignment.

Fifty contractors is roughly two hundred transactions a month, each of which can be wrong. At that volume, manual handling stops being viable and errors stop being visible, because nobody is checking two hundred of anything.

The growth constraint is not usually people. It is that adding contractors adds transactions and working capital simultaneously, which is covered in perm versus contract desks.

The four functions

Timesheets. Collection, approval and chasing. The weekly tax on somebody's time, and every invoice behind a missing timesheet is delayed.

Invoicing. Correct rates, correct periods, client-specific formats. Rate errors are the most expensive failure here because they persist.

Collections. Ageing, queries and escalation. Frequently the largest recoverable cash in the business and the least worked.

Statutory compliance. Provident fund, insurance, filings. Where a lapse becomes a client problem via principal employer liability, per contract labour compliance.

Build or outsource

The decision is usually framed as cost per transaction, which is the wrong frame.

Outsource when volume is low, compliance is standard, and you would be hiring one person who becomes a single point of failure. A payroll provider absorbs regulatory change you would otherwise track yourself.

Build when volume is high enough to justify a team, when client-specific invoicing requirements are heavy, or when the data matters operationally. That last point is underweighted: a provider holding your timesheet and billing data means margin by desk requires an export and a reconciliation.

Hybrid is common and workable: statutory filings outsourced, timesheets and invoicing in-house where the operational data lives.

Why the data should sit with the front office

The most consequential decision is not who does the work but where the record lives.

When placements sit in one system and billing in another, somebody reconciles them monthly, and margin by desk becomes a report rather than a number you can read. Firms in that position cannot answer basic questions quickly: which desk is profitable, which client is worth the effort, what a recruiter actually earned the firm.

That is a commercial cost rather than an administrative one, and it is the argument in where back office margin leaks for one data model rather than two systems and a reconciliation.

Collections, and why recruiters should see it

Finance owns the invoice and has no relationship with the client. The recruiter has the relationship and never sees the ageing.

So an invoice sits for ninety days on a query one call would resolve, because the person who could make that call does not know it exists. Most overdue staffing invoices are administrative rather than solvency problems, and administrative problems respond well to being seen by the right person.

Putting invoice status on the same record as the placement fixes the visibility half. It does not fix a client who cannot pay, and it is worth being clear about which problem you have.

What to build first

Timesheet automation. Reminders, escalation and visibility of what is outstanding before the billing run rather than during it. Highest return per hour of effort.

Rate validation. A check that catches a wrong rate at entry rather than in month four, since this is the error that persists.

Ageing visible to the relationship owner. Cheap, and it usually collects more in a fortnight than a credit control policy change.

Margin by desk. Once the first three work, this becomes readable rather than assembled, and it changes how the firm is managed.

Frequently asked questions

The two reports to run this week

Every invoice over sixty days, and whether anybody has spoken to the client about it. Then every active assignment, and whether the billed rate matches the agreed rate.

The first usually collects cash within a fortnight. The second usually finds at least one error that has been running for months, and nobody was going to find it otherwise.