Client Profitability for Accounting Firms: Which Clients Actually Make You Money

You find your unprofitable clients by estimating what each one costs you to serve, subtracting that from what they pay, and ranking the whole list by margin instead of by fee. You do not need a time tracker to do it. A rough cost per client, ranked once a quarter, tells you more than a perfect number nobody ever gets around to calculating.

That is the whole method. The rest of this guide shows you how to run it in one sitting, what usually turns up at the bottom of the list, and what to do with each client you find there. It includes a fully worked example of a five client firm so you can see the numbers move.

Why does revenue per client mislead you?

Because your biggest clients feel like your best clients, and those are not the same thing.

A client paying a large monthly fee gets attention, gets priority, and gets the benefit of the doubt. Nobody looks hard at how many hours sit behind that fee, because the fee itself looks like proof the relationship is working. Meanwhile a smaller client with clean books and a quiet inbox might be the most profitable account you have, and nobody notices that either.

Revenue tells you who pays you the most. It does not tell you who leaves you the most. That gap is the same trap covered in why revenue is vanity and profit is sanity, just measured one client at a time.

The fix is not better bookkeeping inside your own firm. It is a different question. Stop asking "how much does this client pay?" and start asking "how much is left after we serve them?"

What is the simplest way to estimate cost per client?

You need three numbers, and none of them require perfect data.

1. Your cost per hour

Add up what it costs to run your firm for a month: salaries, payroll taxes, benefits, rent, software, insurance. Divide that by the hours your team actually spends on client work in a month. Not hours at the office. Hours on client files.

If your monthly running cost is $26,000 and your team spends about 400 hours a month on client work, your cost per hour is $65 (example figures, used through the rest of this guide).

That single rate already carries your overhead. You do not need a separate allocation exercise. Every hour a client uses brings its share of rent and software with it.

2. Estimated hours per client

This is where firms without time tracking usually stop, and they should not. Sit down with whoever does the work and estimate, client by client, how many hours a month each one really takes. Include the parts nobody logs: the email threads, the chasing for documents, the review, the "quick call" that ran long.

Then run one sanity check. Add up every estimate and compare the total to the hours your team actually spent on client work last month. If the list comes up well short, the missing hours did not vanish. They are hiding inside a few clients, and they are almost always the noisy ones. Go back and adjust those first.

The estimate will be rough. That is fine. You are not billing from it. You are sorting with it, and a rough sort still puts the right clients at the top and bottom.

3. What each client actually pays

Use what reaches your bank account, not what you invoiced. A client who pays late, disputes invoices, or gets regular write offs costs you more than the fee suggests.

Now the math is simple. Multiply estimated hours by your cost per hour. Subtract that from the fee. Divide what is left by the fee. That is the client's margin.

How do you rank clients by margin, not size?

Put every client in one table, sort by margin from highest to lowest, and look at the order. Here is a fictional firm with five clients so you can see it work. Every figure below is illustrative, not a benchmark.

Look at what the ranking does. The construction company pays the largest fee on the list and sits in the middle (example figures above). The restaurant group is the second biggest client by revenue, and it loses money every month it stays (example figures above). The medical practice, the smallest fee but one, is the most profitable relationship the firm has (example figures above).

Sorted by fee, this firm would protect the restaurant group and worry about losing it. Sorted by margin, it can see the restaurant group is the client it can least afford to keep as is.

Here is the reasoning behind each call:

• Medical practice: keep. Clean inputs, a fee that fits the work. The useful question is what made this client easy, so you can go find more of them.

• Retail shop: keep. A healthy margin on a solid fee. Watch it at the next review, and leave it alone.

• Construction company: restructure. The fee is fine. The hours are the problem. A firm in this position usually finds late documents, job costing questions, and month end scrambles. Fix the process first, with deadlines and a late input fee, and price any real extra work separately.

• Nonprofit: reprice. The hours are reasonable. The fee is simply too low for the work. This one needs a new number, not a new process.

• Restaurant group: reprice or release. Losing money on every month of service. Offer a fee that covers the work. If the client says no, letting them go improves the firm's profit the same day.

What patterns usually show up at the bottom of the list?

When firms run this for the first time, the clients at the bottom tend to fall into a few familiar shapes.

Legacy pricing. Clients who signed years ago at a rate you would never quote today. The work grew, the fee stayed. Nobody raised it because the relationship felt good.

Scope that drifted. A second entity, payroll added, a standing weekly question. The engagement grew one "can you just" at a time. If this looks familiar, the guide to scope creep in accounting firms walks through how to measure it and what to say.

Messy inputs. Receipts in a shoebox, bank statements on the 28th, a bookkeeper on their side who changes every year. The scope never changed. The effort did.

Underpriced from day one. A client you took on when the firm was hungry, at a price that only made sense if the work stayed tiny. It never stays tiny.

The high maintenance relationship. Long calls, urgent requests, and constant second guessing. These clients rarely look bad on a revenue report, and they almost always look bad on a margin ranking.

Notice what is missing from that list: bad people. A client at the bottom is usually a pricing or process problem, not a character flaw. That is good news, because pricing and process are things you control.

What do you do with each unprofitable client?

Every client below your margin floor gets one of three calls: reprice, restructure, or release. Pick one on purpose. Doing nothing is also a call, and it is the expensive one.

Reprice

The work is fine and the fee is wrong. Raise it. Give notice, explain what the engagement includes now compared with when it started, and offer a clear new fee. Practitioners quoted in the Journal of Accountancy on letting clients go put it plainly: when a client takes more time than you bill, tell them the fee is going up and why. Some will leave on their own. The ones who stay are now paying fairly for the work. Either outcome is better than the one you have today. The guide to pricing bookkeeping services covers how to build fees that hold up.

Restructure

The fee might be right, but the way the work arrives is wrong. Set document deadlines. Add a late input fee to the engagement letter. Move recurring extras into a higher package. Push routine questions into a scheduled monthly call instead of an open inbox. You are not charging the client more for the same thing. You are stopping the same thing from costing you more.

Release

Some clients cannot be repriced or restructured into profit, or will not accept either. Letting them go is a business decision, not a failure. The AICPA's Private Companies Practice Section has long encouraged firms to grade every client on factors like pricing, timing, stress, and risk, then rank them A through D and work the D clients off the list, as described in Letting Go: Evaluating and Firing Clients in the Journal of Accountancy. Do it cleanly: written notice, a clear end date, and a tidy handover of their files.

One rule for all three. Talk to the client before you decide they are a lost cause. Plenty of clients at the bottom of the list will accept a fair new fee once they see what the work actually involves. They just never had a reason to look.

Where does client profitability sit in your Financial Management Matrix?

Client profitability is one view inside the Gadal Financial Management Matrix, the part of the 6 Profit Pillars that turns your numbers into decisions. Firm level margin tells you whether the business is healthy. Client level margin tells you why, and which relationships to fix first.

It also connects to everything else you do on profit. Pricing, scope control, and capacity all show up in the same ranking. When you increase profit margins without raising prices, the client ranking is where you see it land. You can find the full framework on the 6 Profit Pillars page, and how it applies to firms like yours on the page for CPA and accounting firms.

Put the ranking on the calendar once a quarter. It takes an afternoon. Each pass gets faster, because the estimates from last quarter give you a starting point.

FAQ

How do I estimate client profitability without time tracking?

Estimate hours per client in one sitting with the people who do the work, including email, chasing, and review. Check the total against the hours your team actually worked on client files last month, and adjust the noisiest clients first if the total comes up short. Multiply by your cost per hour and subtract from the fee. A rough estimate is enough to rank clients, and ranking is the point.

How often should I review client profitability?

Once a quarter is a good rhythm for most small firms. It is frequent enough to catch scope drift before it becomes the new normal, and spaced out enough that it never turns into a project. Also run it before any fee increase or renewal season, so you raise prices where the ranking says to.

Should overhead be allocated to each client?

Yes, but keep it simple. Build overhead into your cost per hour by dividing your total monthly running cost, including rent and software, by client work hours. Then every hour a client uses carries its share automatically. Detailed allocation models rarely change which clients land at the bottom.

What margin should a healthy client produce?

Set your own floor from your firm's target, not from someone else's benchmark. If you want the firm to keep a certain share of revenue as profit, a client well below that share is pulling the average down, and a client below zero is costing you money to keep. Pick your floor, write it down, and review every client under it.

The next step

Your best client might be your smallest. Your biggest might be costing you money. You will not know until you rank them.

If you want a second set of eyes on the numbers, book a free strategy session and we will run a client profitability review on your firm together.