How many subscription clients can one designer actually handle, and can you afford to hire the next one? There is no universal number, but there is a calculation you can do in twenty minutes. This page walks through it with a worked example you can copy.
You need three inputs: the design hours a person really has in a week, the average size of a request, and how many requests each client sends you. Everything else follows from those.
Start with the design hours you actually have
A 40-hour week is not 40 design hours. Before anyone opens Figma, the week has already lost time to stand-ups, Slack threads, briefing calls, file handovers and fixing last week's revisions. Leave and bank holidays take another slice, even in weeks when nobody is off, because the work they displace has to land somewhere.
Here is an illustrative breakdown for one full-time designer. Your numbers will differ. The point is to write them down instead of assuming 40.
- Contracted hours: 40
- Admin, internal meetings and tool upkeep: minus 4
- Client communication, including briefing calls and Slack replies: minus 3
- Buffer for revisions and restarting paused work: minus 3
- Leave and bank holidays averaged across the year (about 26 days): minus 4
That leaves about 26 production hours. If you are a solo founder, you are also the salesperson and the bookkeeper, so take off another three to five hours for proposals and invoicing. Working alone, that probably puts you nearer 22.
Count your own hours in the stack too. A founder who designs 20 hours a week on top of running the studio is doing unpaid production work. If the numbers only work because those hours cost nothing, the offer is not working yet. You have just not sent yourself the invoice.
Turn hours into requests, then requests into clients
Next, you need an average request size. Pull the last month of finished cards from Trello, or whatever board you use, and sort them into rough families. For example:
- Social set of five to eight posts: 2 to 3 hours
- Email header and banner set: 2 to 4 hours
- Landing page design: 6 to 10 hours
- Sales deck of 15 to 20 slides: 8 to 12 hours
Until you have your own history, work from the families above: small requests are more common than big ones, so a mix that averages 3 to 6 hours is a reasonable starting assumption, not a benchmark. If you have no history yet, start at 4 and correct it after a month.
Now a common misunderstanding. "One active request per client" limits how many things you work on for a client at the same moment. It does not limit how much they send you in a week. A client who approves a social set on Tuesday morning can have the next card in your queue by lunchtime. So the number that matters is delivered requests per client per week, not active requests.
| Step | Typical mix | Heavier mix |
|---|---|---|
| Contracted hours per week | 40 | 40 |
| Hours removed (admin 4 + client communication 3 + revision buffer 3 + leave 4) | minus 14 | minus 14 |
| Production hours left | 26 | 26 |
| Average hours per request | 4 | 6 |
| Requests per week (26 ÷ average) | 6.5 | 4.3 |
| Clients per designer at one delivered request per client per week | 6.5, so plan for 6 | 4.3, so plan for 4 |
| Clients per designer at two delivered requests per client per week | 3.25, so plan for 3 | 2.2, so plan for 2 |
Round down, always. Half a client does not exist, and the spare half-slot is what absorbs a bad week.
Look at how much the last two rows move. The same designer carries six clients or three depending on how fast clients approve and send the next thing. That is why a client count on its own tells you very little. Two studios with identical headcount can be in very different places: one comfortably busy, the other drowning. If the second row is your reality, the fix is often in your service scope and pricing rather than in hiring.
What real founders report
The most quoted public numbers come from an Indie Hackers AMA posted on 18 May 2021 by Brett Williams, who runs DesignJoy. At the time he reported 30+ clients and $50-60k in monthly recurring revenue, as a solo designer. Clients could only have one active request at a time, managed through Trello. He said he completed 12 to 13 requests per day, working in two-day sprints, for roughly 50 to 60 hours per week. At that point he was also holding a full-time job.
He did not state a per-client price in the thread. Commenters guessed at one, but those are guesses, so we do not repeat them.
DesignJoy's own site now lists its Monthly Club at $4,995 a month, a limited-time discount from $5,995. It offers one request at a time and an average 48-hour delivery.
Why is 12 to 13 requests a day so far from the 6.5 a week in the table? The inputs are different in every direction. He was working 50 to 60 hours, not 26 production hours. His requests must have averaged around an hour or less for that daily count to be possible. And one very fast, very practised founder is not a model for a studio with employees. Use his numbers as proof that the model can work solo, not as a capacity target.
You will also want churn figures, and you will not find them. Studios do not publish churn or retention numbers, and we will not invent one.
The quiet-client trap
Here is what it looks like on a Tuesday morning. Two clients have gone silent on requests you delivered last week. Their cards sit in "Waiting on client". Your designer has a light day, and it looks like you have room for a new client.
You do not. Each of those waiting requests still owes you a restart. When feedback lands, the designer has to reopen the file and get back into the brief before changing anything. Feedback often arrives in clumps, so three quiet clients can wake up in the same afternoon.
Count it like this. Keep waiting requests visible on the board, never archived. Give each one a restart estimate, usually 1 to 2 hours. Add those hours to the week they are most likely to come back. If the total regularly exceeds your revision buffer, your buffer is too small or your request queue rules need a limit on how long a request can sit waiting before it closes.
Can you afford the next designer?
Start with what a designer really costs you per month, not their salary. Here is the sum as a formula, with example numbers you should replace with your own.
Fully loaded monthly cost = (annual salary ÷ 12) + employer's National Insurance and pension + software and equipment + recruitment and management time spread over the year.
Example only: salary of £38,000 ÷ 12 = £3,167 ($4,196). Add £550 ($729) employer's National Insurance and pension, £150 ($199) software and equipment, and £300 ($398) recruitment and management time. Total: £4,167 a month (about $5,521). Dollar figures use 1 USD = 0.7547 GBP, the mid-market rate on 23 September 2026.
Then divide by your monthly price per client and round up. At £3,000 ($3,975) a client, £4,167 ÷ £3,000 = 1.39, so the designer needs two clients just to cover their own cost. From the worked example, they can carry about six at a typical mix. The other four are what pays for your own time and the studio's overheads, and only once they are actually signed.
These are planning figures, so check any money decision with your accountant before you commit to it.
Before you post the job, ask two questions.
Do you have that many clients waiting? A new designer earns their keep only when their slots are full. If you do not have a waitlist, or current clients already asking for more than you can deliver, you are hiring on hope. Fill the slots first by selling, or by moving the founder's own design hours onto the new work.
Who absorbs the first two months? A new designer is slower while they learn your clients and files, and someone senior reviews everything they ship. In the example that is roughly £8,334 (about $11,043) of cost before they reach full speed, plus your review hours. Make sure you have that cash, and that your own week can take the review load.
Do this in twenty minutes
Open the pricing and capacity worksheet and work through it with last month's board open next to it.
- Write each person's contracted hours, then take off admin, client communication, a revision buffer and averaged leave. Include your own hours if you design.
- Pull last month's finished requests and sort them into three or four families with an hour range for each.
- Work out your average request size. If you have no history, use 4 hours and mark it as a guess.
- Count delivered requests per client last month and divide by four to get a weekly figure for each client.
- Divide production hours by average request size, then by requests per client per week. Round down.
- Add up every request in "Waiting on client" and give each a restart estimate. Check it fits inside your buffer.
- If you are thinking about hiring, fill in the fully loaded cost and the break-even client count, then answer the two questions above.
If step one is guesswork because nobody logs time, look at time and capacity tools, but try a week of rough notes on the board first. Run the sum again next month with real numbers, and compare it with this one.