Section 1
Start after the setup
The instinct is to establish context first. Context is what the reader has not agreed to give you time for yet. So start where the interesting part starts, usually one or two sentences later than feels comfortable, and let the context arrive inside the story rather than before it. A specific claim, an unexpected number, a named mistake: any of these does the work that a preamble cannot. The corollary is that the piece must then deliver on the opening, since an opening that overpromises is the fastest way to train an audience to skip you, and that training is difficult to reverse. Fifteen seconds of context in a video is not an introduction, it is where the audience leaves. The same compression pressure exists in a fundraising conversation, and [Telling Your Story to Investors: What Works, What Doesn't](/blog/telling-your-story-to-investors-what-works-what-doesn-t) covers what survives it there.
Section 2
Serialise instead of summarising
One long post covering everything you know about a topic performs worse than six posts, each covering one thing properly, and the reason is not attention span. It is that a complete summary gives the reader nothing to return for, while a specific piece of a larger picture builds an expectation. Serialisation also matches how these platforms distribute. Any single post reaches a fraction of the people who follow you, so treating each one as if it were the only thing they will see is realistic rather than repetitive. Say the same thing several ways over a quarter. The version you are tired of is the version most of your audience is encountering for the first time.
Section 3
Choosing what to publish
Work through the model that follows before you write anything. It sorts what you know into three columns: things only you have seen, things anyone in your field could say, and things you are repeating from elsewhere. Publish from the first column and use the third sparingly.
Section 4
Reading the response correctly
The visible metrics are the least informative ones. A post can collect a large number of approving reactions from people who will never buy anything, which feels like a result and is closer to noise. The signals worth tracking are narrower: replies that describe the reader's own situation in detail, shares into private groups, and the messages that arrive a week later saying someone was sent this. Those indicate the story travelled through a retelling, which is the only mechanism that reaches beyond your own following. Track them manually if you have to. Fifty entries in a spreadsheet, each recording what was published and what came back, will teach you more about your audience than any dashboard, because the dashboard optimises for the platform's business rather than yours. Formats that invite the reader to participate rather than watch are covered in [Interactive Storytelling: Engaging Audiences in Digital Spaces](/blog/interactive-storytelling-engaging-audiences-in-digital-spaces).
Section 5
The costs nobody puts in the plan
Reach here is rented, and the landlord changes the terms without notice. Building a business on distribution you do not control is a real risk, and the mitigation is boring: move the people who respond onto a list you own, early and repeatedly. The second cost is drift in what you say. Optimising for engagement gradually pulls a founder toward stronger claims, more conflict and less qualification, because those perform, and after a year the public position no longer matches the operator's actual views. That is a reputational liability with your buyers, who are a different and quieter audience. Scheduling and volume tooling makes both problems easier to hit, and [Automating Social Media Management With AI](/blog/automating-social-media-management-with-ai) is worth reading alongside the question of whether more output is the goal.