What works differently on OnlyFans
Starting an agency follows a similar path everywhere: you need creators, a team for the chat, and a calculation that adds up. What that looks like on the German-speaking platforms is covered in Starting a Maloum agency and Starting a 4Based agency. On OnlyFans, three differences come on top, and they do not change the order of the steps, only how much each one weighs.
From your first creator you work in the OnlyFans CRM, which is free during the beta.
The platform sends you no traffic. OnlyFans has no meaningful discovery function. Nobody browses there and stumbles across one of your creators. Every single fan has to come from the outside, Reddit, X, TikTok, Instagram. On Maloum, accounts also grow without external reach; here that is ruled out.
The market is saturated. You are not starting into a growing market but into one with a clear oversupply of agencies. Creators receive enquiries every week. Your first task is therefore not outreach, but a reason why someone should pick you instead of the other twenty.
Your audience does not sleep. The fans are mostly not in your time zone and they write in English. That shifts your staffing needs from day one, not from your third creator onwards.
Your core service is not the chat
This is the statement that decides whether your agency holds up. On a platform with its own discovery, the chat really is the core: fans arrive by themselves and somebody has to turn them into revenue. On OnlyFans, nobody arrives by themselves.
Anyone starting a pure chat agency there is selling creators the second half of their problem and leaving the first half with them. The creator notices by month two at the latest, he pays a share and still has to generate reach himself. That is exactly where most new agencies fail: not at the chat, but at never having had an answer to "where do the fans come from".
Answer that question before you approach your first creator. And answer it with channels, frequency and ownership, not with "we do marketing". Concretely: which platforms get posted on, how often, who cuts the clips, who writes the copy, and which part the creator delivers. If you cannot say that in three sentences, you are starting too early.
Building a chat team before it is settled who brings in the fans. Chatters without traffic are fixed costs without a counterpart, and a first agency almost never dies from bad chatting, but from empty inboxes that still have to be paid for.
The calculation to run before you start
Work out once what actually reaches you out of one euro a fan spends. Few people do this beforehand, and it explains most failed agencies.
OnlyFans keeps around 20 percent. So out of 100 euros, roughly 80 stay in the system. From that, the creator's share is deducted, depending on the arrangement, the larger part. What remains is yours, and out of that remainder you pay everything: chatter hours, content editing, posting on the traffic channels, your own time, tools.
The part that surprises most people: staffing hours are the biggest item, not the platform fee. Covering an account around the clock means several shifts, and those hours are incurred before the revenue arrives. Budget for a lead time, not for things moving in step.
A practical test before founding: take a realistic monthly revenue for a single creator, subtract the platform fee and the creator's share, and divide the rest by the hours you and your team put in. If the resulting hourly rate is one you would not work for yourself, then either the calculation is wrong or the creator is too small, and both are better clarified now than after three months.
Staffing: why you need more people sooner
On a German-language platform you can chat yourself at the start and close up in the evening. On OnlyFans that does not work, because the strongest buying hours for a US audience fall in the middle of your night. Anyone covering only European office hours misses them, and does not notice, because what never happened shows up in no statistic.
What that means in practice
You need a second person for evening and night coverage sooner than expected, and that person has to be able to write in English, not through a translator, but in tone. Fans do not notice the difference in a single word, but in the sense that something feels off, and then they simply stop replying.
The shift nobody volunteers for
If you pay your chatters purely on commission, the night shift becomes a problem: the weakest hours are where the least is earned. Whoever wants coverage has to pay for it, through a base rate for that shift or a higher percentage. How to set that up cleanly is covered in Paying chatters: base, commission or both.
Start with one creator, not five
The reflex when founding is to have several accounts as fast as possible, more accounts, more revenue. In practice that is the quickest route to being mediocre at all five at once.
With one creator you learn what works on this platform: which traffic channels carry which niche, how often you have to post, which approach sells in chat, when the strongest buying hours fall. That knowledge transfers. Without it, the second creator only multiplies your uncertainty.
The honest side effect: a single satisfied creator is your best outreach. In a market where everyone gets cold enquiries weekly, a recommendation from someone who genuinely earns more than before counts for more than any self-presentation.
The system, before the second creator arrives
As long as you manage one account, you hold everything in your head. From the second that stops working, and from the third it costs money: which chatter has which buy rate, where did this fan stall, what was this fan promised three weeks ago. Reconstructing that from group chats and notes loses exactly the information that turns conversations into revenue.
The usual route is a spreadsheet, and it works surprisingly long, up to the point where several people are supposed to work in it at the same time. When the switch pays off and how you notice it is covered in Why standard tools fail.
Build the system before you need it.
CRM and translator for OnlyFans: with buy rate per chatter, fan history and every creator in one view. The bot runs on Maloum and 4Based. Everything is free during the beta: start now and the processes are in place from day one.
Frequently asked questions
What do you need to start an OnlyFans agency?
Above all a solid answer to the question of where the fans are supposed to come from. OnlyFans has no meaningful discovery function, so all traffic has to be generated externally. Anyone who only builds a chat team is selling creators the second half of their problem. Everything else, team, tools, processes, comes after that.
What does an OnlyFans agency actually earn?
There is no blanket answer, but the calculation is clear: the platform keeps around 20 percent, the creator's share comes off that, and out of the remainder you pay chatter hours, content editing, traffic building and your own time. The largest item is staffing hours, not the platform fee, and those are incurred before the revenue arrives.
How many creators should I manage at the start?
One. With a single creator you learn which traffic channels carry which niche, how often you have to post and when the strongest buying hours fall. That knowledge transfers. Without it, the second creator only multiplies your uncertainty, and a satisfied creator is also your best outreach.
Why do I need more staff sooner on OnlyFans?
Because the audience is international and the strongest buying hours for a US audience fall in the middle of your night. Covering only European office hours misses them without anyone noticing, what never happened shows up in no statistic. So you need a second person sooner, one who can write English in tone, not just in words.
Is an agency worth it on OnlyFans despite the many providers?
The market has a clear oversupply and creators receive enquiries every week. That does not rule out founding one, but it changes the order: your first task is not outreach, but a demonstrable reason why someone should pick you over the other twenty. As a rule that reason is provable traffic competence, not a lower split.
When do I need software instead of a spreadsheet?
As long as you manage one account, your head is enough. From the second it gets messy, from the third it costs money, because nobody can say any more which chatter has which buy rate and what a fan was promised three weeks ago. The tipping point is not the number of creators but the moment several people are supposed to work on the same data at once.
