For practices · The payroll decision

Should your practice run payroll in-house or refer it out?

Most small accountancy practices did not decide to offer payroll. It arrived. A good client asked, somebody said yes, and eight years later it is a fixed cost with a person attached to it. This is the decision nobody sits down and makes, set out plainly.

There are only three options, and one of them is not really available

Run it in-house. Refer it away to a bureau and take a commission or nothing at all. Or run it under your own brand and have somebody else do the work. Those are the three. The fourth option, quietly stopping, is not available to most firms, because payroll is the service that keeps a client talking to you twelve times a year instead of once.

That frequency is the whole argument for keeping it, and it is a good argument. It is also the reason a practice tolerates payroll long after it has stopped paying for itself.

Start with the arithmetic, not the principle

Take one payroll client with five employees, paid monthly. Count the real time: collecting the hours, chasing the two people who never send them, running it, checking it, issuing payslips, filing the FPS, answering one query, and dealing with a starter or a leaver every few months.

Most firms land somewhere between forty minutes and an hour and a half a month once the queries are in. Then price your own time honestly, at what you would charge for advisory work rather than at a notional cost. For a lot of practices the answer is that payroll runs at a loss and is subsidised by the year-end fee.

That is not an argument for dropping it. It is an argument for knowing which of the three options you are actually choosing.

When keeping it in-house is right

Keep it if payroll is a genuine part of what you sell and you have the capacity to do it properly. Keep it if your client base is concentrated enough that one person can hold the whole thing in their head. Keep it if you are big enough that a dedicated payroll person is busy all month rather than busy for four days and idle for the rest.

And keep it if the alternative would leave you unable to answer a client question without picking up the phone to somebody else. That last one matters more than firms expect.

When referring it away costs you more than it saves

Referring a client to a payroll company gets the work off your desk and puts a third party between you and your client. The commission, where there is one, rarely covers what you have given up. You lose the monthly contact, you lose the early warning when a client starts struggling, and you inherit the complaint when the other firm gets it wrong without any of the control that would have let you prevent it.

Referral makes sense when payroll is genuinely peripheral to your practice and you are comfortable with the client having two relationships instead of one.

The third option, and what it actually looks like

The middle route is to keep the client relationship and stop doing the work. The payroll runs under your brand, the payslips carry your name, your client never deals with anyone else, and the processing happens somewhere else.

Practically, that means you send the changes each month and approve the run. Everything after that, the filing, the payslips, the year end, the auto-enrolment duties and the HMRC correspondence, sits with the bureau. Your client sees the same service and the same firm.

The two questions worth asking before you go near this are simple. Who is liable when it goes wrong, and what happens if you want to bring it back in-house next year. A bureau that answers both without hedging is worth talking to.

Three tests that settle it faster than a spreadsheet

First, the holiday test. If your payroll person took three weeks off in April, what happens? If the answer is uncomfortable, you do not have a payroll department, you have a person, and that is a risk rather than a service.

Second, the growth test. If you won four payroll clients next quarter, could you take them? If the honest answer is no, then payroll is capping the rest of your practice, because those four clients came attached to year-end work you also cannot take.

Third, the margin test. Look at what you bill for payroll against what you bill that client for everything else. If payroll is ten per cent of the fee and forty per cent of the contact, you already know which way this goes.

What we do, in one paragraph

We are a UK payroll bureau and we run payroll wholesale for accountancy practices, under your brand. You keep the client, you keep the fee you charge today, and you decide what to charge. We do not contact your clients and we do not sell to them. The wholesale rates are on the partners page, along with what is and is not included.

If you are weighing this up, the useful first conversation is not about price. It is about how many payslips you run, how much of the month they take, and what you would do with that time instead.