My Agency Is Taking Too Much — Renegotiate or Leave?

Last updated August 21, 2026 · by the Perlage Studios team

If you’re earning well and the split has started to feel wrong, you have two moves: renegotiate the deal or leave it. Both start in the same place: knowing your real number. Both go badly when they start anywhere else. This guide is the decision path: how to work out what your agency actually takes, what to check in the contract before you open your mouth, how to run the negotiation, how to read their reaction, and when the dip that comes with leaving is still the cheaper option. It’s written for creators already inside an agency deal, and none of it is legal advice.

How do I work out what my agency is actually taking?

Take your last three payout periods and divide what actually reached your bank by what the account earned after OnlyFans’ 20%. That is your effective share, and it is the only number worth negotiating over. If it doesn’t match the percentage in your contract, start there.

The contract number and the real number are rarely the same thing. For each of the three periods, lay out three figures: what the account earned after the platform kept its 20% of all creator earnings, what the agency’s share came to, and what got deducted on top (tool fees, chatter costs passed through, chargebacks, anything labelled “expenses”). Then divide what landed in your bank by the first figure. Three periods, not one, because a single month hides one-offs in both directions.

One thing shifts that number more than it sounds: whether your split is calculated before or after the platform’s 20%. The base changes your real share, and we’ve broken that math down separately.

For context on the ugly end of the market: a BBC investigation that heard from 60 UK-based creators found, as reported by Eastern Eye and IBTimes UK, that contracts reviewed by the BBC commonly took around 50% and in some cases up to 70% of earnings, on top of the 20% OnlyFans had already kept. That is a warning marker, not a market rate. What a given percentage should actually include in services is a separate question, covered in our guide on splits and fees. And if you want the full cost picture at higher revenue, that math has its own guide — here, all you need are your own three payouts.

Is the split the problem, or is the revenue the problem?

They need different conversations. If revenue is growing and the work is visible, you are negotiating whether the percentage matches the scope. If revenue is flat or falling, a better split just slows the bleed. That is a performance conversation, and it comes first, because terms follow output.

Run your effective share against the revenue trend before you book anything. An agency taking a defensible percentage of a shrinking number is a different problem from an agency taking an indefensible percentage of a growing one, and mixing the two conversations is how negotiations go sideways: you walk in about money, they answer about effort, nobody leaves with anything signed.

If the real issue is output (dead campaigns, slower chat, sliding PPV numbers), you’re in the underperformance guide’s territory, and its section on raising problems before you quit covers that conversation. This guide is the other lane: the work is fine, the terms aren’t. Be honest about which one you’re in, because asking for a lower split from an agency that’s underdelivering treats the symptom and reprices the disease.

While you’re valuing the service, it’s worth knowing what you’re actually buying. For most agencies the core product is the chat operation, and that model is under legal scrutiny: subscribers have brought a class action against OnlyFans’ operating companies (Brunner v. Fenix Internet, LLC and Fenix International Limited, U.S. District Court, Northern District of Illinois) over agency chatters posing as creators, and a similar class action filed in July 2024 has been ordered to trial in 2027. None of that decides your split. It just belongs in the picture of what the percentage pays for.

What do I check in my contract before I say a single word?

Three clauses: your termination window, auto-renewal, and penalty clauses. The moment you ask for a better split, you show your hand — so know before you speak whether your notice window is about to close, whether the contract renews itself, and what leaving would cost you on paper.

Read the contract before the conversation, not after, because the ask itself changes your position: once you’ve raised the split, they know you’re re-evaluating the deal, and any clause that rewards them for waiting starts working against you. The full anatomy — splits, terms, ownership — is in our contracts guide; for the negotiation you need exactly three points.

First, the termination window, and second, auto-renewal — the two travel together. BuzzFeed News documented contracts from Los Angeles agencies in 2021 that renewed automatically for three years or more unless the creator gave written notice inside a window ranging from two weeks to two months. So find both dates: when your window opens, and what the contract rolls into if it closes on you. If the window is coming up, put it in your calendar before you send a single message. Being able to leave on schedule is your entire leverage, and losing it mid-negotiation turns you from a counterpart into an audience.

Third, penalty clauses. The same reporting found heavy penalty clauses in reviewed contracts: a five-million-dollar penalty for confidentiality violations, a hundred thousand dollars per breach of a non-compete. It also found one creator who was asked for around $400,000 on her way out. In the contracts the BBC reviewed, IBTimes UK reported, one creator was told leaving would cost her £10,000. You’re not reading these clauses to panic. You’re reading them to know whether you negotiate from safety or from inside a trap — and if it’s the trap, a lawyer reads them next, before you send anything at all.

How do I renegotiate the split without blowing up the relationship?

Bring your three-payout calculation, ask for one specific change, and put it in writing. Name the number or the scope adjustment you want, give them a date to respond, and don’t threaten to leave unless you mean it. A bluff they call once costs you the whole negotiation.

Open with the calculation, not the feeling. “I’ve run my last three payouts and my effective share comes out at X while the contract says Y” is a sentence an agency has to answer. “I feel like I’m paying too much” is a sentence they can absorb. Then make one concrete ask. That can be a different percentage, but it doesn’t have to be: a scope adjustment works too — if you already run your own socials, stop paying for social management — and so does a lower share above a revenue level you’ve already proven you can hold.

Keep the whole exchange in writing, give a response date, and treat a verbal yes as a no until it’s a signed amendment with an effective date. Watch the counter-offer, too: a better split traded against a longer term is a lock-in with better lighting, not a discount. And don’t put leaving on the table unless you’re ready to walk through the door. A called bluff ends the negotiation and sours whatever deal survives it. One caveat: nobody publishes reliable data on how these talks go. This section is our own experience, nothing more.

What does my agency’s reaction tell me?

A professional counterpart treats the question as normal business: numbers on the table, a counter-offer, a written amendment. Stalling, guilt, vague promises with no date, or anger at being asked are all data. The reaction to a fair question is the cheapest audit of an agency you will ever run.

The good version is boring. They defend their number with their own math, they counter, someone drafts an amendment. Disagreement about the percentage is fine — that’s what a negotiation is. What you’re watching is how they treat the fact that you asked.

The warning signs are stalling past your response date, guilt as an argument, promises that never get an effective date, or a sudden change in temperature. The far end of that spectrum is documented: in the BBC’s investigation, as Eastern Eye reported, one creator described receiving repeated warnings that she would “get what’s coming to you” after trying to renegotiate her manager’s percentage. Most reactions won’t be that. But the pattern scales down, and a milder version carries the same information: this counterparty punishes the question. An agency that treats a fair ask as betrayal has just told you how the exit will go. Weigh that in the next section, and if anything resembling a threat shows up, skip straight to the one after it.

When is leaving the better deal, even with the dip?

When the numbers still favor leaving after an honest handicap for the switch: revenue usually softens for two to four weeks after a move. If your effective share is far from fair, the relationship failed the reaction test, or the contract punishes you for asking, the dip is the cheaper option.

The comparison is your current effective share against a realistic alternative — another agency at honest scope, or going solo and buying back only the services you actually use — minus a handicap for the transition. We’ll say the uncomfortable part plainly, because it’s our own market: switching costs revenue at first, usually for two to four weeks. Budget for one softer month so the dip can’t pressure you into signing whatever ends it fastest. The week-by-week handover and how to manage that dip are the switch guide’s job, so we won’t repeat them here.

Leaving wins when the gap is structural: an effective share the scope can’t justify, a counterparty that failed the reaction test, a contract that made asking feel dangerous. Renegotiating wins when the relationship is functional and the number is the only thing wrong. A good agency on bad terms is a fixable situation, and fixing it is cheaper than moving. If it is the exit, the mechanics — written notice, backups, the order you take your logins back — are in the leave guide, and if the deal you’re leaving did real damage, start with the recovery guide instead.

What if they threaten me or claim I can’t leave?

Stop negotiating and start documenting. Save every message, check who holds your logins and payout details, and take the contract to a lawyer before you reply. An agency claiming you can’t leave is making a legal claim, and legal claims get tested on paper, not in a chat thread.

First, understand what tends to be at stake. In the BBC’s reporting, via Eastern Eye, several creators said managers had changed passwords, controlled account settings or restricted access to their own profiles, and IBTimes UK noted managers being handed logins and access to payout routes. If any of that describes your setup, the logins come before any negotiation. We’ve covered who should hold what and what to do about a lockout separately.

Ownership claims have a public paper trail. BuzzFeed News reported in March 2022 on two influencers who sued their management company and said that when they decided to leave, the agency claimed ownership of their OnlyFans accounts and threatened to sue them in return. The underlying litigation is public record: Elizabeth Machabeli & Jane Doe v. Unruly Agency LLC and others, Los Angeles Superior Court, Case No. 21STCV41395, filed November 2021, with related cases 21STCV31028 and 21STCV26060 — everything in those complaints being, at this stage, allegations. The takeaway isn’t any verdict. It’s that “we own your account” and “you can’t leave” are claims that end up in front of judges, not settled facts you have to accept in a DM.

For a sense of how seriously this end of the spectrum is taken: when the BBC put its findings to the UK’s Independent Anti-Slavery Commissioner, Eleanor Lyons, she said the accounts described recognised indicators of exploitation — coercion, financial pressure, and restrictions on a person’s ability to leave an arrangement freely. That is the sober reason “just stay and keep your head down” stops being a neutral option once threats are in the room: a deal you’re afraid to leave isn’t a deal anymore. Save everything with dates visible, stop arguing in the chat, and spend the money on an hour with a contracts lawyer. Nothing on this page is legal advice, and this is the point where that stops being a formality.

How does Perlage handle splits and exits?

Revenue share only, calculated on net earnings after OnlyFans’ 20%, with no setup fee, no exit fee, and a month-to-month contract. There is no fixed percentage on the website because the fair number depends on the scope of work, so the split is agreed openly in the strategy call.

The month-to-month part matters most for this guide’s question, because it changes what renegotiation is. When either side can end the deal at the end of any month, with no notice period, the split has to keep earning its place — there is no lock-in doing the persuading, and no penalty clause doing the retaining. That’s also why there’s nothing to pay on the way out: we think the open door is what keeps the work honest.

And we’re not neutral here — an agency wrote this page. Which is exactly why the useful next step is one that doesn’t require trusting us.