Where the capacity goes in an accounting firm

In a lot of professional services firms every growth question resolves to the same answer, which is another hire. Not because hiring is the best lever, but because headcount is the only lever anyone has ever measured. Revenue grows, overhead grows alongside it, and the margin never widens.

The ceiling that follows is real and most owners can feel it well before they can describe it. What makes it hard to act on is that the alternative levers — utilisation, cycle time, work mix — are all ratios, and ratios do not appear on a P&L.

Take your senior headcount. Multiply by the gap between your current utilisation rate and the rate you believe is achievable. That difference is capacity you are already paying for.

On 30 senior staff at 67% utilisation against a 76% target, that is a 9-point gap — about 2.7 full-time equivalents, or roughly 100 billable hours a week that currently produce nothing.

Be careful about what that is worth, because this is where the number is usually oversold. At a A$250 average charge-out across 45 working weeks it is a bit over A$1.1 million of billable capacity. But recovered capacity is not recovered revenue. It converts only to the extent there is demand waiting to fill it, and the honest version of the figure is always discounted for that — sometimes heavily. A firm with a waiting list and a firm with a soft pipeline get very different answers from identical arithmetic.

The harder term is the target. The 76% used here is an input, not a standard we are asserting about your firm. What it should be depends on your work mix, your leverage model, and how much genuinely non-billable partner work your seniors carry. Establishing a defensible target is part of the assessment, because a gap measured against the wrong target is not a business case.

The second number is simpler and more often ignored. Work is completed, and then it sits. If your average is 49 days from matter close to invoice, that is seven weeks of finished work being funded by the firm.

The arithmetic is direct: annual billings divided by 365, multiplied by the lag, is roughly the cash tied up in completed-but-unbilled work at any moment. On A$28 million of billings and a 49-day lag, that is about A$3.8 million.

None of that requires AI to see. It requires knowing the number. Most firms can quote their debtor days from memory and cannot quote their matter-to-bill lag — which is unfortunate, because debtor days are largely the client’s behaviour and the billing lag is entirely your own.

Practice management software has done billing reminders for twenty years. If reminders were the answer, the lag would already be gone.

What is usually missing is configuration against reality — thresholds set per partner and per matter type rather than on one generic schedule, so the prompt lands at a moment that matches how that person actually works. Whether that is worth doing in your firm depends on a prior question: is your lag a process problem or a behaviour problem? They look identical in the data and they need completely different fixes, and answering it wrongly is how firms end up with an automation nobody uses.

Run against 12 months of WIP records, billing data and timesheets. Measure utilisation and matter-to-bill lag directly rather than estimating them. Establish a defensible target for each from your work mix rather than from a benchmark table. Separate the process half of the lag from the behavioural half. Then rank what surfaces and pick one thing.

If the numbers come back small, we say so. A firm already running 78% utilisation and billing inside a fortnight has a materially weaker case than the one modelled above — and it is better for everyone that this shows up in week one than after an implementation.

It does not tell you your numbers. Every figure above is arithmetic run on illustrative inputs, chosen because they are plausible — not because they were measured at a particular firm. No client engagement underlies any of it.

The inputs that matter most are the ones you would have to supply: your utilisation rate, a defensible target for it, your billing lag, and an honest view of how much recovered capacity your pipeline could actually absorb. Change any of them and the case changes with it. That is the honest position, and it is why the first step here is a measurement rather than a proposal.