How many clients can a bookkeeper handle: a capacity model, not a magic number

Ask ten bookkeepers how many clients one person can handle and you will hear numbers from 10 to 60, each stated with confidence. They are all describing their own practice. The honest answer is that “clients” is the wrong unit. A sole trader with one bank account and forty transactions a month is not the same load as a restaurant with payroll, three cards, and a weekly supplier run. Capacity is measured in hours, and the number of clients is what falls out once you know the hours each one takes. This guide gives the formula, a worked example to replace with your own figures, what published surveys do and do not show, and the levers that change the answer. Sources are listed in the Sources section.

What the surveys show

The Institute of Certified NZ Bookkeepers publishes an annual benchmark survey. Its 2023 edition, with 145 completed responses, charts total active clients per respondent in fifteen bands from “1 to 5” to “201 +” ^[1] . Two things stand out.

First, the distribution is nearly flat. No band held more than 12% of respondents. There is no hump in the middle that could be called typical.

Second, the figure is per business, not per person. The same survey shows that most respondents engage employees, contractors, or both ^[1] , so a business reporting 150 clients may be six people.

The survey also records what respondents ranked as their most pressing challenge. Hours of work has been at or near the top in each year reported, and recruiting experienced staff, offered as an option for the first time in 2023, was chosen by 21% ^[1] . The profession’s own data says capacity is the constraint. It does not say what the right number is.

Be careful with any article that states an average number of clients per bookkeeper without naming a survey. We could not find one that traces to published data.

The capacity formula

Clients you can carry = (client hours available per month × (1 - buffer)) ÷ average hours per client per month

Three inputs, each of which you can measure.

Client hours available

Start from the hours you work and subtract everything that is not client work: your own admin and billing, sales and onboarding calls, training, software problems, and leave averaged across the year. For most solo practitioners the result is well below the hours worked. Use your real diary for the last two months, not an intention.

Buffer

Bookkeeping work is not even through the month. Month-end, quarter-end returns, and year end stack on top of each other, and something always goes wrong for someone. Hold back a share of capacity for peaks. A practice with every client on the same quarter needs a larger buffer than one whose clients are staggered.

Average hours per client

This is the number that decides everything, and the only way to know it is to track time by client for a month or two. Estimates are consistently too low, because they leave out the query emails, the chasing, and the call that was supposed to take five minutes.

A worked example

The figures below are illustrative inputs, not benchmarks. Replace every one with your own.

A solo bookkeeper works 150 hours a month and spends 40 of them on admin, sales, and training. That leaves 110 client hours. With a 15% buffer, 93.5 hours are plannable.

No real practice is all one type. With a mix of 10 light, 8 standard, and 2 heavy clients, the load is (10 × 2) + (8 × 5) + (2 × 12) = 84 hours, leaving 9.5 hours of headroom: room for one standard client and a light one, or nothing at all if a heavy client is the next to call.

The table makes the point the opening made in words. “How many clients” has a sixfold range inside one person’s month, depending only on who the clients are.

What drives hours per client

When two clients on the same fee take very different time, the cause is almost always in this list:

• Transaction volume, across all bank, card, and payment accounts.
• Document quality and timing. A client whose documents arrive complete costs a fraction of one whose documents are chased. See how to stop chasing client receipts.
• Number of accounts and entities. Each bank account is a reconciliation; each entity is a close.
• Payroll, especially with variable hours.
• Filing frequency. Monthly returns triple the deadlines of quarterly ones.
• Responsiveness. Queries answered in a day against queries answered in two weeks.
• Industry quirks. Construction retentions, hospitality tips and stock, ecommerce payouts.
• State of the books at takeover. A file that started as a clean-up stays expensive unless the clean-up was finished. See catch-up bookkeeping.

Rank your clients by tracked hours divided by fee. The bottom few are where your capacity is going.

Levers that change the answer

Capacity is not fixed. Each lever below lowers hours per client or raises the hours available.

Standardise the close. One sequence for every client, in the same order, with the same checklist. Switching between bespoke routines is a hidden cost. The sequence is in the month-end close process.

Invest in technology, continuously. In the 2024 CPA.com and AICPA PCPS survey of 206 US client advisory practices, about half (51%) said they continually invest in technology, and that group reported serving 50% more clients than all respondents, 100 against 67 ^[2] . The figures describe practices, not individuals, and an association, not proof of cause, but the direction is consistent with what the formula predicts: fewer hours per client, more clients per practice.

Concentrate on a niche. The same survey found practices with more than half their revenue from defined industry niches reported 38% higher median revenue from client advisory work than all respondents ^[2] . Familiar clients take less thinking: the same suppliers, the same tax traps, the same questions.

Fix onboarding. A client set up properly, with access, a document route, and a calendar from day one, costs less every month afterwards. See the client onboarding checklist.

Price for the hours. If a client’s fee does not cover their tracked hours, either the fee rises or the client leaves. Both restore capacity. See pricing bookkeeping services.

Delegate before you are full. The same survey reported median net client fees of US$156,250 per professional for 2023 ^[2] , a reminder that growth in a healthy practice comes from adding people as well as adding clients per person. For scale on the cost side, the US Bureau of Labor Statistics puts median pay for bookkeeping, accounting, and auditing clerks at US$24.36 an hour for 2025 ^[3] .

Capacity for a team

The formula works for a practice as well as a person, with two changes.

First, run it per role. A junior who processes and a senior who reviews have different client hours and different work, so one combined figure hides the real bottleneck. In most small practices that bottleneck is review: the owner checks everything, so the practice can take on only as many clients as the owner has hours to review, however many people are processing.

Second, count the cost of coordination. Every client handled by two people needs a handover, a review, and questions passed back. Allow for it in hours per client, or the team’s capacity will come out lower than the sum of its members.

The practical test is the same as for one person: track hours by client and by role for a month, then look at who runs out first.

Signs you are past capacity

• Closes finishing after the day you promised.
• Returns filed in the final 48 hours as a habit, not an exception.
• Your queries to clients going out late, so their answers arrive late.
• Review steps skipped because there is no time.
• New clients onboarded with “just send me what you have”.
• No week in the quarter without a deadline.

Two of these together mean the answer to “can I take one more?” is no, until hours per client come down.

Where ExpenseFlow fits

ExpenseFlow acts on one input in the formula: hours per client spent on bills and receipts. Documents arrive by upload, by a per-client email address, or from Google Drive. Each is read line by line and given an account and tax treatment, with a confidence score, and held for review. The bookkeeper’s part becomes checking and approving, and corrections are learned per client so the same supplier is coded the same way next time. Approved items sync to Xero or QuickBooks Online.

ExpenseFlow does not publish a time-saving figure, and this guide will not invent one. The effect depends on how much of a client’s hours are document handling in the first place. Track hours per client for a month before and a month after, and use your own number. On cost, billing is per client file with unlimited staff seats, so adding a team member does not add software cost; see pricing for bookkeepers.