When to Pitch a Sale to a Fan: Message Rank Decides

Most of the argument about fan chatting is about wording: which line, which tone, which teaser. The thing we can measure across a corpus this size is cruder and more useful. It is not the sentence. It is where in the conversation the sentence lands.

Across 42,852 conversations, 1.5 million messages and 38,879 sales, one pattern holds harder than anything else we have looked at: a pitch placed before the sixth message converts roughly a third worse than the same pitch placed later, and the best-performing position sits after about ten exchanges.

When should you pitch a sale to a fan?

After about ten exchanges, and never before the sixth. That is the entire rule.

Read it as three instructions:

• A floor. Below the sixth message you pay a measured penalty of roughly a third of your conversion rate. No wording recovers it.
• A target. Around the tenth exchange is where offers do best across the corpus.
• Permission, not a trigger. Reaching message ten does not oblige anyone to pitch. It removes the reason not to.

The floor is the part that pays, because it is the one crossed thousands of times a week without anyone logging it.

What actually counts as one exchange?

A turn the fan took. A chatter who sends five messages into silence has produced volume, not depth, and a thread counted that way will get pitched early by someone convinced they waited.

Counting rules that survive contact with a real inbox:

• Consecutive messages from one side count once. Three messages in a row from you are one turn, not three.
• A one-word reply is a turn, but a weak one. It keeps the thread alive; it does not say the fan is ready.
• Scheduled sends do not count. A broadcast the fan never answered has not moved the conversation anywhere.
• A reopened thread starts again. After weeks of silence the fan is not at exchange eleven in their own head. Our measurement does not settle that case; treat it as a fresh warming phase.

What the fan has given you changes sharply with depth, and that is the real reason position matters:

Depth of the thread	What the fan has usually given you	What an offer there is asking of them
First message	A click and a greeting	Trust a stranger with their card
Under six messages	A name, a mood, some politeness	Buy on the strength of nothing
Sixth to about the tenth	A reason for being there, a preference	Answer a priced version of what they raised
After about ten	Questions of their own, specifics, time spent	Confirm something they already want

The columns explain the number. An offer before the sixth message is not badly written. It is addressed to someone who has not yet told you anything you could price.

Why does an early pitch cost so much?

Because it spends a card that only gets played once. Four mechanisms stack up:

• A refusal is not neutral. Once a fan has said no, every later offer is a re-offer, and a re-offer carries the memory of the first one. The early pitch does not just fail. It raises the price of the next attempt.
• It re-labels the account. A fan sold to in the first minute learns what the inbox is for. Their replies get shorter, and short replies are where threads go to die.
• It pre-empts the fan’s own request. The cheapest sale in any inbox is the one the fan asks for. Pitch first and you never find out what they would have asked for.
• It hides the good threads. A chatter who pitches everything early converts a few fans who would have bought anyway, and burns the rest. On a dashboard that looks like effort.

What does the rule change for a chatter mid-shift?

It replaces instinct with two checks: the count, and one signal from the fan. Both have to be true before the offer goes out. Signals that genuinely mean a thread is ready:

• The fan asks a question of their own, about the creator or about what is available.
• They volunteer something specific and unprompted: a preference, a schedule, a reason.
• They answer in full sentences rather than acknowledgements.
• They come back the next day without being nudged.

Things that feel like signals and are not: a fast reply, a compliment, a run of one-word answers, or a fan who is simply online a lot. Speed is not readiness. The cleanest form of the rule is a floor plus a signal, never a message number on its own. That is how fan chatting scripts writes the rule into a script.

How do you build a shift plan around message rank?

By making depth visible in the queue and making the handover carry the count. A shift plan that ignores depth guarantees the same thread gets pitched twice by two people, both of them at exchange one as far as they can tell. Four things a handover has to carry per thread:

Field	Why the count depends on it
Exchanges so far	The counter must not restart with the chatter
Offer made, and at what depth	Prevents a second pitch presented as a first
What the fan asked for	The signal that permits the offer at all
What was refused, and at what price	Stops a re-offer reading as a fresh start

Two scheduling decisions follow. Threads past the tenth exchange get finished by whoever is on, not parked for a named owner who returns tomorrow. Depth is perishable. And a shift ending mid-warming hands over a note, not just a thread; the mechanics sit in how to run a chatting agency. Position in the thread is only one of the two timing questions. The other is the clock, and it works differently (see what time do fans buy).

Does your pay scheme push chatters to pitch early?

Almost certainly, if it rewards the attempt rather than the outcome. This is the part most agencies never connect to their conversion numbers, and it overrides every script ever written.

Pay basis	What it maximises	What it does to message rank
Per message sent	Volume of text	Pads threads with filler, then pitches to justify the volume
Per offer or per PPV sent	Number of attempts	Pitches early and repeatedly; the worst structure available
Daily target in offers sent	Attempts before the shift ends	Pitches cluster at the end of a shift, regardless of depth
Commission on revenue	Closed sales	Broadly aligned, but pushes towards discounting under pressure
Hourly, no variable part	Presence	No pressure either way; needs real quality control
Hourly plus a bonus on conversion	Well-placed offers	Closest to aligned; needs enough threads to read honestly

Two conclusions. Never pay for a pitch as an event: it is the one line item that directly buys you the measured penalty. And track the average depth at which each chatter makes their first offer: one number per person, minutes to pull, and more revealing than any transcript review. It is also worth testing before you hire, which how to hire chatters covers.

Can you pitch too late?

Yes, and it is a real failure mode rather than a rhetorical balance. Our corpus shows an optimum after about ten exchanges, not a curve that keeps rising. A chatter who is never comfortable pitching produces warm, pleasant, unprofitable threads, and every exchange past readiness is one not spent on another fan. If the fan has asked for something specific, the count stops mattering. Sell.

How do you check this on your own threads?

Pull the depth of the first offer in each thread, bucket it, and compare conversion between the buckets. It is a two-column question: rank of the first pitch, did it convert.

• Use threads from one period, so a change of price grid does not explain the gap.
• Split at the sixth message first, since that is where the effect is largest.
• Run it long enough that one high-spending fan cannot decide the answer.
• Then read the same number per chatter, not only per account.

If your split disagrees with ours, trust yours and keep measuring. If you have never looked, start with the floor: no offer before the sixth message, and a signal before any offer at all. That one rule moves more than a rewritten script does.

The protocol itself (how many conversations per variant, what you measure, and what you forbid yourself from changing at the same time) is set out in testing what works in conversations.