9.2 Numbers You Can Trust, and Numbers You Cannot
Every revenue figure in this field belongs to one of three buckets: numbers collected first-hand from operators, numbers a platform publishes about itself, and numbers that exist to sell you something. Mixing the three is how an operator ends up budgeting against a month that never arrives. This chapter separates them, runs the arithmetic that reconciles a platform's headline revenue with what its median creator takes home, and gives a six-question test you can apply to any revenue claim before it is allowed to shape a plan.
[AS OF 2026-10] Platform scale, commission rates, and creator policy change quarterly. Re-check each figure against the platform's current terms before relying on it.
Three kinds of number
| Bucket | Example | What it is good for | What it is not |
|---|---|---|---|
| First-hand operator figures | $4,230 in a month from one account | Setting expectations for a working account | A forecast for your account |
| Platform-published figures | A platform's annual recurring revenue and creator count | Sizing the market and the median | Your income |
| Vendor and course-seller figures | A claimed $300k in 90 days | Nothing, until a denominator appears | Evidence of anything |
The buckets are not ranked by honesty. A platform figure can be perfectly accurate and still useless as a personal forecast, because it describes an average across a population with a very long tail. A course seller's figure can be impossible to check.
The two accounts we can check, and the distribution
These are the numbers this book treats as its floor of evidence, because they come from operators who run pipelines like the one in Part VII, and because the composition of the revenue is stated.
Account one, May 2026. $4,230 in the month, approximately $9,173 lifetime. Composition: roughly two-thirds from chat, subscriptions about one-seventh of the total. [AS OF 2026-10]
Account one, August 2026. Best month above $8,000, lifetime approximately $25,000. [AS OF 2026-10]
Account two. $20,000 lifetime, more than 80% of it from chat. This operator treats the subscription as a quality filter rather than as the product.
The arithmetic is worth doing, because it shows the shape. In May, lifetime stood at $9,173 and a single month produced $4,230 — so everything before that month summed to $4,943. Three months later, lifetime had risen to about $25,000, a gain of roughly $15,800 across June, July, and August. With the best of those months above $8,000, the other two averaged a little under $4,000 each.
What that tells you. The account was still ramping, not plateaued. Revenue was uneven month to month — a peak of $8,000 plus, sitting on top of months near $4,000 — and the lifetime figure was dominated by recent months rather than accumulated slowly. If you had read only the August headline, you would have concluded the account earned $8,000 every month.
What the distribution actually looks like. Most accounts earn between $0 and $500 a month. A smaller active group reaches $1,000 to $10,000 in its first year. [UNVERIFIED] The band is compiled from practitioner reporting across forums and operator conversations rather than from audited platform data, and it is consistent with the platform arithmetic in the next section.
The distribution is a long tail with a thin shoulder, not a bell curve. The mean sits far above the median, which means the average is not a description of a typical account. A plan built on the average will be wrong in one direction: it will assume revenue arrives sooner and larger than it does.
The practical consequence. Budget the pipeline from chapter 7.5 against the $0–500 band, and treat any month above it as information about what worked rather than as a new baseline.
Reading the platform's own numbers
Fanvue's published position, as of the research period: $200 million in annual recurring revenue, roughly 325,000 creators, a 20% commission, and identity verification for the operator behind the account. [VENDOR CLAIM] [AS OF 2026-10] The scale figure comes from the platform, which has an interest in it being large; the commission and the verification requirement are operational terms you can check in the signup flow.
Now do the arithmetic, because it is the most useful calculation in this chapter. Two readings are possible and both are instructive.
| If the $200M figure means | Creator payouts | Per creator, per year | Per creator, per month |
|---|---|---|---|
| Gross volume processed (creators keep 80%) | $160M | ~$492 | ~$41 |
| The platform's own 20% cut | $800M | ~$2,462 | ~$205 |
Neither reading supports a $4,000 month. The first-hand operator in the previous section sits an order of magnitude above the average creator on the platform that hosts this business. That is not a contradiction; it is what a long tail looks like from the inside, and it is the best available confirmation that the $0–500/month band is not pessimistic.
One more reading of the same number. Annual recurring revenue is a platform metric. A platform adding creators quickly will show revenue growth even when median creator earnings fall. A rising platform number and a falling median are compatible.
And a policy note that belongs with the figures. OnlyFans bans synthetic personas; Fanvue is the subscription platform that permits them. [AS OF 2026-10] A permitted category is a policy decision, and policy decisions get revisited. Treat platform permission as a renewable agreement with a date on it, not as a property of your business.
The headline benchmarks
Two figures circulate as proof that this business works at the top.
- Emily Pellegrini, reported at roughly $23,000 a month. [AS OF 2026-10]
- neuro-sama, reported at roughly $166,000 a month from subscriptions. [AS OF 2026-10]
What they are. Existence proofs. They answer one question — is there real money at the top of this market? — with a yes, and that answer is worth something when you are deciding whether to start.
What they are not. Forecasts, and the second one is not even the same business. neuro-sama is a live streaming persona with years of audience, real-time interaction, and a subscription product built around a show that runs on a schedule. Its revenue does not generalize to an account posting short-form video, and presenting it as a comparable is a category error.
How to use them. As a ceiling argument, when a brand asks whether synthetic personas can carry real revenue. Keep them out of your own planning spreadsheet.
A revenue figure without its denominator is a headline, not data. The three numbers that make a claim checkable are the number of accounts it covers, the time window it measures, and whether the figure is gross or net.
Numbers that are marketing
These surface constantly in forums, on YouTube, and in the descriptions under both.
[UNVERIFIED] $300,000 in 90 days from a single persona. No verifiable source, no dashboard with context, and the claim appears in the funnel of a paid course or a template pack.
[UNVERIFIED] Agencies reporting $300,000 to $400,000 a month. Again unsourced. Where a number this size is real, it usually describes the total revenue of a company running dozens of personas and several operators, not the income of one account.
Why they are not evidence, stated plainly.
- The claim carries a referral link or a course. The source earns when you believe the number.
- Revenue is presented where profit belongs. Commission takes 20%, payment processing takes a slice, and chat hours are unpaid labor. Net is a fraction of gross.
- The denominator is missing. One account or two hundred? One operator or a team of six?
- The window is a launch, not a track record. Ninety days is the period in which a funnel either works or does not; it is not a year.
- The failures do not post. The accounts that earned nothing do not publish retrospectives, so the visible sample is the survivors.
The one question that dissolves most of them: who gets paid if I believe this? If the answer is the person making the claim, the number needs independent corroboration before it enters your plan.
How to read any revenue claim
A procedure you can run in five minutes on any figure, including your own.
- Name the source's interest. Course, tool, agency, Promoting someone else's product for a commission: the cheapest first dollar, the weakest position. program, or nothing but reputation.
- Convert gross to net. Subtract the platform commission — 20% on the platform named above — then payment processing, then refunds.
- Convert revenue to profit. Deduct the pipeline cost from 7.5 and the hours spent in chat. A $4,000 month and a $900 pipeline cost is a $3,100 month minus your time.
- Demand the denominator. How many accounts, how many operators, how long.
- Check the window. Best month, median month, or cumulative?
- Ask what happened to the accounts that failed. If nobody knows, the sample is selected.
Scoring it. A claim that fails three or more of those six tests is marketing, whatever the number is. A claim that passes five is a data point you can plan against.
The template for an honest claim, which is what this book tries to meet: here is the month, here is the composition of the revenue, here is what the pipeline cost, and here is how long the account has existed.
What this changes in your plan
Three adjustments, applied to the pipeline and revenue plan from Part VII.
Budget on the band, not the headline. Model the pipeline at the $0–500 level and let it scale when the numbers arrive. This is the same posture 7.5 takes toward generation tiers: buy capacity slightly ahead of demonstrated consumption.
Measure composition, not only totals. The first-hand accounts earn two-thirds to four-fifths of their money from chat. If your revenue is arriving from a different place — affiliate or a A paid deliverable for a company, quoted as units with usage rights and exclusivity written in. — you are running a different business than the one the benchmarks describe.
Write down your denominator now. Account count, months live, gross, and net. In three months, that single line is the only thing that will let you compare your own numbers against any claim in this chapter.
Where this goes wrong
- Planning against a best month. The August figure above is a peak sitting on months near half its size.
- Reading a platform's total revenue as creator income. The arithmetic puts the average creator between roughly $40 and $205 a month.
- Repeating an unsourced claim because it is exciting. The referral link in the description is the tell.
- Comparing gross revenue to net profit. Commission, processing, and your chat hours all come out of the same figure.
- Treating neuro-sama as a comparable. A live subscription show is a different product from a posting account.
- Ignoring the failures. Survivorship makes the visible sample look like the whole distribution.
- Assuming platform permission is permanent. A policy decision with a date on it, not a property of your business.
- Never writing down your own denominator. Then no claim, including your own, is checkable.
The gate
- Can you state your own gross, net, account count, and months live on one line?
- Which bucket does every number in your plan come from?
- Have you subtracted commission, processing, and your hours from every revenue figure you use?
- Do you know which stream produces most of your revenue, and does it match the first-hand composition?
- Which headline figure is currently shaping your expectations, and does it pass the six-question test?
- If your best month doubled next month, would you treat it as a baseline or as information?
What you leave with
Three buckets for every number in this field, with first-hand operator figures separated from platform-published scale and from course-seller claims. Two checkable accounts, one with $4,230 in a month and roughly $9,173 lifetime on about two-thirds chat revenue, and the arithmetic that shows the distribution underneath the headlines. A reconciliation of a $200 million platform figure against per-creator earnings, the two news benchmarks with their limits stated, a six-question test for any claim, and one line of your own numbers to measure everything else against.