fan types

Fan types are a written list of behaviours, each with a plain rule for what to send. Every account already sorts its audience somehow, even if the sorting only exists in the head of whoever is on shift; writing it down is what turns that instinct into fan types. The nearest term is fan segmentation, which does something else. Segmentation cuts the base into groups so that a mass message reaches the right people. A fan type describes one person inside one conversation, so the writer knows the next move without reading six months of history.

Which types of fans actually change what you write?

Four, and each is defined by what the fan has done rather than by who he says he is.

Type	What defines him	What it changes
New	Subscribed, has not bought yet	Nothing is pitched yet; the exchange is still being built
Regular buyer	Buys, at a familiar rung	The price tier is known, so the offer is quick
Whale	Spends out of proportion to everyone else	Top of the ladder, made-to-order work, never a discount
Expired	Subscription lapsed, account still reachable	A win-back, not a sales message

What the type sets is rhythm and price level. It does not set the moment of the pitch: our data shows that pitching before the sixth message drops conversion by roughly a third whoever the fan is, and the optimum sits after about ten real exchanges. No category buys you the right to go early.

Which distinctions change nothing?

Most of the ones people are tempted to record. The fantasy he mentioned once, whether he writes in short bursts or long paragraphs, the pet name he has for the creator: all of that can be useful inside a conversation and none of it routes a message.

The test is one question: does the label change what gets sent next? If it does not, it is a note, not a type. Notes belong in the thread, where the next person will read them. Types belong in the short list the whole team works from.

How does a team keep everyone using the same types?

Three or four. A fifth almost always turns out to be a note that got promoted, and it costs you: every extra type is one more decision at the top of every conversation, and one more thing to explain at handover. A list nobody can recite from memory is a list nobody uses.

Two habits keep it usable:

• One rule per type, written as an instruction rather than a description: “offer the middle rung”, not “moderate spender”.
• Merge rather than split. Two types that produce the same opening line are one type.

When does a label go stale?

Fast, and that is the failure mode nobody plans for. A type is a snapshot of recent behaviour: a regular buyer who has not opened anything in two months is no longer a regular buyer, and treating him as one produces offers that miss.

A type describes a period, not a person. Re-read it whenever the conversation resumes, and let the fan’s last actions overrule whatever the record says. The point of the list was never to file people. It was to spare the next writer the scroll back through months of thread.