A second opinion on your agency — without quitting first

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

You earn well, you have a team, and something is nagging at you. Or nothing is, and you’d just like proof before the next renewal that the numbers are what they should be. Either way, the instinct to get an outside read before doing anything drastic is the right one. You don’t have to quit, threaten to quit, or sit through three sales pitches to find out where you stand. A second opinion is one person outside the deal reading your numbers. This guide covers what that look can see, what it can’t, who counts as independent, and what to do with the verdict. If revenue has already dropped and you want the cause first, start with the drop diagnosis; if you want the do-it-yourself symptom checklist, that lives in the underperforming guide. Nothing here is legal advice.

What can an outside reviewer actually check without your login?

More than you might expect. Everything you can screenshot or export yourself — earnings history, subscriber trends, pricing, posting cadence, your public feed, your contract — is enough for a serious first read. A reviewer works from what you hand over. Nobody needs your password to form a view, and anyone who asks for it first should worry you.

The package you hand over decides the quality of the read, and all of it is already in your hands: earnings month by month, as far back as your own statistics go; subscriber counts over the same stretch; how much of the money arrives through messages rather than subscriptions, however your statements break it out; your subscription price and whatever promo logic runs underneath it; and the agreement you signed. Add whatever the agency sends you (reports, plans, the occasional invoice), because the gap between what they report and what the account shows is itself a finding.

Any reviewer worth listening to starts by re-deriving what the deal costs you. OnlyFans takes a 20% commission on creator earnings before anyone else gets paid, and whether your agency computes its share before or after that platform cut changes your real number by more than most people expect. The arithmetic is in our guide to the cut before or after the 20%. A second opinion should run it on your actual statements, not on the percentage printed in the contract.

The public layer needs no documents at all: posting cadence, caption quality, the pricing visible on the page, how often the profile runs discounts. Anyone can read it, which is exactly why it belongs in the review: it’s the storefront your agency runs in your name.

What a first read does not need is your password. Screenshots and exports carry every number that matters at this stage, and handing a stranger your login to settle an argument with your agency would be trading one access problem for another. The reviewers who insist on account access before any conversation are answering the independence question for you.

Do you have to tell your agency you’re getting a second opinion?

No. Reviewing your own numbers and paperwork with an adviser you chose is normal professional behavior, not a betrayal, and you don’t need permission or an announcement. Check your contract for a confidentiality clause before forwarding the document itself, and let a lawyer read anything unclear. Whether you say something afterwards depends on what the review finds.

Agencies review their own performance constantly — that’s what their internal reporting is. You’re allowed the same professionalism about them. An agency that’s doing its job survives scrutiny without ever noticing it happened, and one that would take offence at the idea has told you something useful about itself.

Two practical rules make the quiet version work. First, don’t run a loyalty test. Announcing “I’m thinking of leaving” to see how they react starts the endgame before you have the facts, and you can’t un-start it. The conversation with your agency comes after the review, with specifics in hand. How to have it is in the underperforming guide. Second, keep the account running exactly as normal while the review happens. If you change behavior mid-review, the reviewer reads a performance, not the business.

The one caveat sits in the paperwork. Some agreements carry confidentiality clauses, and what they cover varies. Read yours before you forward the document itself to a third party, and if the wording is unclear, spend the hour with a lawyer first. That hour is cheap insurance. Again, nothing in this guide is legal advice.

What should a second opinion look at that you can’t see yourself?

The work done in your name where you only see the output: above all the chats sent under your identity, then how your pricing and effort compare with what the reviewer has seen working elsewhere, whether your revenue hangs on one lever or several, and whether the trend you have quietly normalized is actually normal.

Start with the chats, because they’re the part of the operation carried out in your voice and read by nobody who answers to you. This isn’t a hypothetical concern. In N.Z. et al. v. Fenix International Limited et al., No. 8:24-cv-01655, a class action filed in federal court in California in July 2024, subscribers allege that management agencies employ paid chatters who impersonate creators in paid messages. A second, similar case was filed in 2025 in the U.S. District Court for the Northern District of Illinois, Case No. 1:25-cv-03244.

On December 12, 2025, U.S. District Judge Fred W. Slaughter dismissed the suit without prejudice, with leave to amend until January 2, 2026. The plaintiffs did amend, and in May 2026 the court let one claim proceed: a privacy claim against the agencies over chatters being given access to creator accounts and fan data (our chatter-liability guide walks through it). The December reasoning still matters: the agencies acted with the creators’ permission, and OnlyFans’ terms of service disclose that creators may use third-party agents. In plain terms, agency-run chatting is disclosed. It is still done under your name. That’s exactly why a second opinion should read a sample of the messages sent under it — not to catch a crime, but to see what your name is being used to promise.

A reviewer reads those transcripts for the things the revenue line hides: promises you’d never make, pressure that burns a loyal spender for one purchase, a tone that doesn’t match your feed, the copy-paste rhythm of a script stretched across too many accounts. You can’t see this yourself for the same reason you can’t proofread your own writing: you’re inside it, and you were never meant to be the audience.

Past the chats, the outside view is good at structure: whether your income hangs on one mass-message lever or several sources, whether growth depends on discounting, whether the work delivered matches the work agreed. And it’s good at trends you’ve normalized. A slow slide looks flat from the inside, because you adjusted to each step down as it happened. If the slide has already happened and you want the cause, the structured diagnosis is in why did my income drop.

What can no second opinion tell you?

Whether you would have earned more with a different team. There is no counterfactual: nobody can rerun your last year under different management and compare. It also cannot verify what your agency does off your account, and it cannot hand you reliable industry benchmarks, because almost none exist in verifiable form. It reads your account against itself.

Be suspicious of anyone who claims otherwise. “With us you’d be making three times this” is a pitch wearing the costume of an analysis; nobody can price a year that didn’t happen. An honest reviewer talks in ranges and in tests — this looks heavy on discounts and thin on retention, here is what I’d try for thirty days — and keeps the facts separate from the interpretation.

Benchmarks deserve the same suspicion. The verifiable numbers in this industry are platform-wide: OnlyFans processed $7.22 billion in gross fan payments in fiscal 2024 and paid out $5.8 billion to creators. Those totals tell you the market is enormous, and nothing about what an account like yours should earn. The creator-level statistics floating through agency marketing are almost never sourced, and a serious review declines to use them. Your account gets compared against its own history, because that’s the only baseline that verifiably exists.

And no outside reviewer can measure the relationship. Whether messages get answered, whether you trust the person on the other end, whether you’ve been talked out of things you were right about — you are the only witness to that. The review gives you facts to put beside the feeling. It can’t replace the feeling, and it shouldn’t try.

Who counts as independent — and who is just pitching you?

Almost nobody, completely. Another agency reviewing your numbers is also prospecting — including us. Genuinely disinterested reviewers exist: a lawyer paid by the hour for the contract, an accountant for the money, a creator at your level for the operations. Use interested reviewers for analysis if you like, but disinterested ones for the decision.

Rank reviewers by what they gain from each possible verdict. Another agency has the sharpest operational eyes in the market and the most obvious motive: their review is also their pre-sales. That doesn’t make their numbers wrong. It means you weigh the verdict knowing who benefits, and it hands you a clean test: a useful agency review contains findings that don’t end in “hire us.” If “stay and renegotiate” isn’t an outcome they can name, you’ve heard a pitch. This applies to us. A strategy call with Perlage is a review by an interested party, and you should hold it to the test in this paragraph.

A lawyer or an accountant paid by the hour is the closest thing to true independence: no stake in whether you stay or go, and a bill that’s the same either way. Their limit is scope: they read documents and money, not marketing. If the deeper question is whether you need an agency at all at your size, that’s covered in is an agency worth it if you earn well.

A creator at or above your level is the underrated option: no motive, current pattern knowledge, and she has seen her own numbers, which is more than most commentators can say. Her limit is a sample size of one. Anonymous forums are the noisiest version of the same thing: occasionally right, unverifiable, and permanent, so never post identifiable numbers there.

Whoever you pick, independence shows in what they ask for. Exports and screenshots: reasonable. Your login, before any conversation has happened: no. A reviewer who can’t form a first view from what you can safely hand over isn’t reviewing — they’re collecting.

What do you do with the verdict — stay, renegotiate, or switch?

Sort every finding into three piles: fine, fixable inside the current deal, and structural. Fixable problems justify a renegotiation with specifics and a deadline attached. Structural ones — the economics, ownership, trust — usually justify a change. A clean report is worth keeping too: date it, and rerun the comparison in six months.

If the review comes back mostly clean, it wasn’t a wasted exercise. It’s the cheapest insurance you’ll buy this year. Write the verdict down with a date on it, file the exports it was based on, and rerun the comparison in six months. Vague unease doesn’t accumulate into a record; dated reviews do.

Fixable findings (chat quality, reporting cadence, effort that drifted) justify a renegotiation, not an exit. Go in with the specifics from the review and a deadline attached, and treat the response as part of the data. What’s actually negotiable lives in the agreement itself; how agency contracts work walks the clauses one by one.

Structural findings are different: economics that stopped making sense at your size, ownership in the wrong hands, chats you would never have approved, trust that didn’t survive the reading. Those rarely improve through conversation. If that’s the verdict, the mechanics are in how to switch agencies: timing, handover, and the dip that comes with any change. And who owns your account and content tells you what to secure before you say anything.

One last honesty: verdicts expire. Accounts move, teams change, and a review from last winter describes a business that no longer exists. Whatever you decide — stay, renegotiate, or switch — decide it on this quarter’s numbers, not on a memory of the year the deal was signed.

Quick answers