Why busy practices aren’t always profitable

Activity is not the same as profit

A firm can be completely full and still make thin margins if the work is inefficient, underpriced or full of rework. More activity on a leaky operation just means more leaks.

Where the profit leaks

• Scope creep — extra work done for free because scope and change control are weak
• Underpricing — fees that haven’t kept pace with cost or value delivered
• Rework and errors — jobs done more than once
• Admin on senior time — expensive people doing cheap tasks
• Poor workflow — time lost to chasing, waiting and switching

Fixing it

Tighten scope and pricing, remove rework, push admin down, and get the workflow flowing. None of that is about effort — it’s about how the firm is run. That’s the work a fractional COO does.

The number that explains it: recovery by job type

Most practices know their overall recovery and almost none know it by job type. That single gap is why the problem persists, because a healthy average conceals the two things you actually need to see.

Work it out for your last twenty jobs of each type: fee billed, divided by time actually recorded at charge-out rates. Then do the harder version — add an honest estimate of the time nobody logged, because unrecorded time is where the worst jobs hide. Preparers under pressure stop recording on the job they know is over budget, which means the file that is losing most money looks the most profitable on the report.

Those four findings need four different responses, and a practice working only from an average will apply one of them to everything.

Scope creep is not the client being difficult

It is the commonest single leak and it is almost always self-inflicted.

A client rings with a question. It takes twenty minutes and it is not billed, because billing twenty minutes feels petty and the relationship matters. Repeat across two hundred clients and a year, and a practice has given away a substantial part of its capacity in increments too small for anyone to notice or object to.

The answer is not to start billing twenty-minute calls, which damages the relationship for very little money. It is to decide what is inside the fee and say so at the start — a fixed allowance of ad-hoc support in the engagement, beyond which work is quoted. Clients accept this readily when it is set out in advance. What they resist, reasonably, is an invoice for something they thought was included.

The version of this that costs most is the client who was priced as straightforward and turned out not to be. That is a scoping failure at onboarding, not a pricing failure, and it is why consistent onboarding pays back in margin rather than just in tidiness.

The tail

Most practices carry a group of clients that cost more to serve than they pay. Everybody suspects who they are; very few have looked.

Rank every client by contribution — fee, less time actually recorded, less an honest estimate of the unrecorded. In a typical practice the bottom fifth contributes very little and consumes a disproportionate share of the disruption, because small clients are not smaller versions of large ones. They generate the same number of queries, deadlines and chases as a client paying five times more.

The right response is rarely a cull. It is to reprice and let them choose. A client who accepts a proper fee has become a different client; one who declines has made the decision without a difficult conversation. Either outcome is better than the status quo, and the capacity released is only worth something if you have decided in advance what it is for — otherwise it fills with more of the same within a month.

Where the money actually goes: a worked shape

Take a practice with £900,000 of fees and twelve people. Suppose rework runs at 8% of delivery time, chasing at 5%, and the tail of unprofitable clients consumes another 5% of capacity at close to nil contribution.

That is roughly a fifth of the practice's productive capacity, none of which appears as a line on any report. It shows up instead as write-offs, as a recovery percentage that is quietly disappointing, as overtime in January, and as the owner's sense that everyone is flat out and the bank balance does not reflect it.

Those percentages are illustrative, not a benchmark — the point of the exercise is to replace them with your own. But the structure holds: in a busy practice that is not making money, the gap is almost never in the fee line and almost always in what happens between winning the work and delivering it.

The order that works

1. Measure recovery by job type, including an honest estimate of unrecorded time. Nothing else on this list can be prioritised without it.
2. Fix process before price where the spread within a job type is wide. Repricing a variable process just moves the loss.
3. Reprice the oldest cohort, which is nearly always the worst, and do it at renewal rather than as an event.
4. Define what is inside the fee and put it in the engagement letter.
5. Deal with the tail last, once you have somewhere for the released capacity to go.

Steps two and three in the wrong order is the commonest mistake, and it is expensive: a practice that raises prices on an inconsistent process keeps the inconsistency and adds a reason for clients to scrutinise it.