How to sell in OnlyFans DMs: where most of the revenue comes from and how a chat team works it

The subscription is the door charge. On well run pages, 60 to 80 percent of revenue comes from what happens after a fan walks through it: pay-per-view sent in a conversation, tips left mid chat, and customs a fan asked for himself. The feed gets him to subscribe. The inbox is where he decides how much he is going to spend with you this month.

This is not a quirk of the platform. It is how the buying works. A fan does not pay five dollars a month because he wants a library of posts. He pays because there is a person on the other side, and he keeps paying when that person remembers him. A feed cannot remember anyone. A DM can, and that is the product.

New and unmanaged pages often sit closer to 30 percent from DMs, not because their fans are different, but because nobody is in the inbox when the fans are. The gap between 30 and 70 percent is not better content. It is coverage, timing and a structure for what gets offered to whom.

A new subscriber is never more curious than in his first hour, and the first ten minutes are the peak of that. He just paid. He is scrolling, he is on the page, and he is deciding whether this was a good idea. If the first thing he hears from you is nothing, he decides on his own, and the answer is usually to scroll on and forget.

The welcome message has to land inside that window and it has to read like a person. His name if you have it, one easy question, one line that sounds like it was written today. No paragraph of rules, no menu of prices. The goal of the first message is his first reply, because a fan who replies once behaves completely differently from a fan who only watches.

Then the welcome offer. Low priced, sent shortly after the opener, positioned as something small for new fans. Its job is not margin. Its job is the first yes: it establishes that buying from you feels good, easy and worth it. A fan who has bought once, even for a few dollars, is a different buyer for the rest of his subscription.

The most common way to lose a fan is to open with a price. A locked message as the first thing in his inbox reads as a vending machine, and fans do not chat with vending machines. They buy from them once, if at all, and then stop opening. Two real exchanges before anything priced is the working rule, and it is not about being coy. It is about being someone he wants to hear from.

The reason is simple. A fan is in one of two modes at any moment. In talk mode he is venting about work, asking about your day, replying quickly to nothing in particular. Pushing a price here costs you the fan. In buy mode he asks what you have, references a post, tips without being asked. Now you name a price fast, because hesitation reads as inexperience.

Reading which mode a conversation is in, and switching your response accordingly, is most of the skill in selling by DM. The best chatters do not sell more often. They sell at the right moment, and they spend the rest of the time making the next right moment more likely.

Price the relationship, not the file. The most common pricing mistake is charging by production cost: longer video, higher number. Fans do not buy production cost. They buy a moment in a relationship with you, and the same clip is worth different amounts to a new subscriber, a warmed up regular and a fan who has bought everything. One price for everyone is two wrong prices out of three.

In practice this becomes a ladder. It starts with the welcome offer, moves into mid priced PPV inside real conversations, adds bundles that make sense against buying alone, then customs priced like made to order work, and at the top offers that exist only for a handful of fans. The order matters because each step is easier to climb once the one below it has been taken.

The ladder also solves the question every creator asks, which is how much to charge. The honest answer is that there is no single number. There is a number for a fan on day one, a number for a fan in week three who has bought twice, and a number for a fan who asked for something nobody else gets. The chatter's job is to know which one he is.

On almost every page, a small share of fans produce a large share of the revenue. Regulars buy the rhythm PPV, tip now and then, and rebill month after month. Whales buy customs, take the top of the ladder, and sometimes spend in one evening what the rest of the inbox spends in a week. Both groups deserve different treatment from the fan who subscribed yesterday, and most pages give them exactly the same.

The first rule is that a whale should never drown in broadcasts. If your best fan receives the same locked message as four hundred strangers, you have told him he is one of four hundred. The more a fan spends, the more personal the offers should get: a line written for him, a reference to what he bought last, something he asked for weeks ago finally made.

The second rule is notes. Name, job, the dog, what he bought, what he asked for and did not get. A fan who feels remembered will quietly outspend a stranger who saw better content. This is not extra effort on top of the business. At this layer of the inbox, memory at scale is the business, and it is the main reason chat teams keep a record on every spender.

Scripts have a place, and it is smaller than most agencies think. The welcome opener, the welcome offer, the structure of a win back before expiry: these can be templated because the situation is the same every time. What cannot be templated is the middle of a conversation, because the fan is not reading a template. He is reading a person, and he notices when the person went missing.

The tell is easy to spot from the fan's side. The same greeting to three hundred fans. A price before he has said a word back. Replies that answer a question he did not ask. Fans forgive a lot, but they do not forgive being processed, and once a fan reads you as a script he stops opening your messages, which means everything you send after that earns nothing.

The working compromise is a playbook, not a script. Situations get a structure, the wording does not. A chatter knows what a welcome flow needs to achieve, what a custom negotiation looks like, how to move a talk mode conversation toward an offer without forcing it. Inside that structure, every message is written for the fan in front of them. That is slower per message and far more profitable per fan.

Mass messages work, and they are also the fastest way to train fans to ignore you. The difference is what surrounds them. A locked message every morning to the whole list, with nothing human between one and the next, turns the inbox into a spam folder. Fans stop opening, then they stop rebilling, and by then the mass message that caused it looks like it was working because a few people bought.

Done properly, a mass message is an event inside a rhythm of real contact. Fans who are in active conversations do not get it at all, because they get something better. The rest get it rarely enough that it still reads as news, with copy that sounds like her and not like a store, and at hours when fans are actually online, which on most pages means late evening in the fan's time zone.

Segment before you send. New subscribers, regulars, whales and expired fans should not receive the same message, and often three of those groups should receive nothing. A good rule of thumb: two real check ins for every priced message a fan sees, whether the priced message came from a chatter or a broadcast.

A fan who messages at 11pm is acting on an impulse that hit at 11pm. Reply inside fifteen minutes and you are inside the mood that made him reach out. Reply at 9am and you are a forgotten notification. Impulse has a half life, and the heaviest buying window on most pages sits between roughly 10pm and 2am in the fan's time zone, which is exactly when a creator alone has the least energy left.

That is the case for coverage as a schedule rather than a mood. The inbox needs to be worked in shifts across the hours fans buy, including the ones nobody wants, and it needs to be worked by people who read the notes. Elvision Studios runs fan chatting around the clock for its creators and reports the numbers weekly, and both halves of that matter: coverage produces the revenue, the report is how you know it did.

A weekly report should be short and honest. Not screenshots of the best conversation, but the numbers that show whether the inbox is doing its job: message revenue split by PPV, tips and customs, how fast fans got a reply, how many new subscribers bought the welcome offer, who the top spenders were and what they bought, and what is planned for next week. If a team cannot produce this, it is not really running the inbox.

Most pages do not lose DM revenue to one big error. They lose it to a few habits that each look harmless and together cap the inbox. The good news is that every one of them is visible in the numbers above, which is a large part of why weekly reporting exists.

The first is blast only PPV: locked messages to the whole list with no conversation around them, which is a spam folder with extra steps. The second is the firesale spiral: dropping prices whenever something does not sell, which teaches fans to wait you out and never pay full price again. The third is ignoring quiet subscribers, the fans who paid and said nothing, who are not lost but simply have not been asked a good question yet.

Behind all three sits the same mistake: treating the inbox as admin instead of as the highest paid hour of the day. Fix the coverage, keep the notes, price by relationship, and let the conversation carry the offer instead of the other way around. None of it requires better content. It requires someone in the inbox while the impulse is still alive.