Instagram

Verified against ChatGPT · 2026-08-11

Write a Broadcast Channel update that reads like an insider note, not a repost of the feed caption

Drafts an Instagram Broadcast Channel message for a creator's closest subscribers, written to feel like a direct, informal update rather than a copy-pasted version of whatever already went out on the main feed.

ChatGPT (GPT-5.1)5 fillable variables

The prompt

Ready to copy — highlighted parts are example details you can swap.

You are writing a message for an Instagram Broadcast Channel — the one-to-many channel a creator sends directly to subscribers who opted in specifically to hear from them, separate from the main feed. The people reading this chose to be here for something more direct than a public caption, so the message has to earn that trust rather than reuse feed copy.

UPDATE TO SHARE
The glaze color that sold out in nine minutes on the public drop is getting a second, smaller restock next Tuesday.

HOW THIS RELATES TO SOMETHING ALREADY POSTED PUBLICLY
The original drop was announced in yesterday's reel, which most of this list has probably already seen.

WHY THIS SPECIFIC GROUP GETS IT FIRST OR DIFFERENTLY
Broadcast subscribers get first access 24 hours before the restock goes on the public grid.

DESIRED SUBSCRIBER RESPONSE
React with which of two restock times works better for them, since we're picking the drop time based on replies.

CREATOR'S VOICE IN DMS
Short, a little breathless, uses 'ok so' as an opener, no corporate phrasing at all.

RULES
Write this in the voice the creator actually uses in a direct message to someone they know, not the more polished, third-person-adjacent voice of a public caption — Broadcast Channel messages that read like a repurposed feed caption are the single fastest way to make subscribers mute the channel, since the entire value proposition of subscribing was hearing something different. State plainly, without over-explaining, why this group is getting this update first, earlier, or in more detail than the general audience — the exclusivity has to be named, not just implied, or it reads as an identical broadcast that happened to arrive in a different inbox. If the update relates to something already posted publicly, reference it briefly rather than re-explaining it in full; assume the reader may or may not have seen the public post, so give enough context to follow along either way without repeating it wholesale. Close with the desired subscriber response stated as something concrete and answerable in the channel's reaction or reply feature — a specific question, a poll-style either/or, or a direct ask — rather than a vague 'let me know what you think.'

WHAT NOT TO DO
Do not add a hashtag block, an emoji-heavy sign-off, or any element that reads like it was written for public discovery — this channel has no discovery mechanic, so anything written to be found by strangers is wasted here and signals the message wasn't actually written for this specific audience.

OUTPUT FORMAT
1. The Broadcast Channel message, ready to send.
2. One line stating what makes this message something a subscriber couldn't have gotten from the public feed alone.

Customize

Optional — swap in your own details for the highlighted parts above.

Why this works

The explicit instruction to write in DM voice rather than caption voice targets a specific and predictable drift: without a strong steer, GPT-5.1 tends to default toward the more polished, slightly promotional register it associates with 'Instagram content' in general, because that's the dominant pattern in its training examples of Instagram copy — but a Broadcast Channel is structurally a direct message thread, and content written in feed register inside a DM thread reads as visibly mismatched to the format the reader is actually looking at, which is exactly the tell that makes a subscriber mute a channel. Requiring the exclusivity reason to be stated plainly rather than implied closes a second gap: an unprompted draft will often just repeat the update without ever explaining why the channel is the right place to hear it, leaving the reader to wonder whether this message is meaningfully different from the public post at all, which erodes the entire reason someone subscribed. The explicit 'what not to do' section banning hashtags and discovery-oriented elements matters because those are reflexive additions the model reaches for on any Instagram-labeled task by default pattern-matching to public-post conventions, and they are actively counterproductive here since a Broadcast Channel has no algorithmic discovery surface for a hashtag to serve — including one signals the message was written generically rather than for this specific, closed audience, which is the opposite of what a channel update is supposed to communicate.

What you get back

Ok so — remember the glaze that sold out in nine minutes yesterday? We're doing a second, smaller restock next Tuesday, and you're hearing it 24 hours before it goes on the grid because you're here. React with 1 if morning works better, 2 if evening does — picking the drop time off whichever wins.

Verified against

ChatGPT GPT-5.1 · 2026-08-11

Changelog

  • 2026-08-11 Initial publish, verified against ChatGPT GPT-5.1.

Need this built into your business?

If a prompt isn't enough — what Scult builds, built and maintained for you — that's Scult's day job.

EXPLORE WHAT SCULT BUILDS
All Instagram prompts

Check your AI visibility

One URL in, a 0–100 score and the exact fixes out.

RUN THE CHECK

Browse all the tools

15 tools across six categories
13 of them never send your data anywhere

Free · No signup · No trial clock

SEE THE DIRECTORY