An essay is an argument.
Build the argument first.

One sentence in the middle of the board: Write what you are arguing as a single sentence a stranger could disagree with. If it cannot be disagreed with, it is a topic, and a topic will drift for twenty minutes and lose the viewer somewhere near minute six.

Branch the objections, not just the supports: Put the strongest counter-argument on the canvas next to your claim. An essay that answers the obvious objection out loud feels rigorous. One that pretends the objection does not exist feels like a pitch.

Kill the version that does not survive: Test two or three framings side by side on the same board and keep the one still standing after you have argued against it. Doing this now costs an afternoon. Doing it in the edit costs the project.

Group sources under the claim they serve: Cluster papers, articles, threads, and clips under the specific point each one proves or damages, so a section is written from a group of cards rather than from memory and a tab you cannot find again.

Keep the link on the card, always: Attach the article, paper, or video URL to the card the moment it lands, which is the only reliable way to have an accurate source list and credit sequence when the video is finished months later.

Clip frames straight from source video: Pull stills out of a YouTube or Vimeo source onto the board when a single frame is the evidence, and the reference stays attached to the image instead of living in your download folder.

Room for a genuinely long project: The canvas is infinite with no object cap, so a research pile that grows for six months stays on one board and stays navigable rather than being split into files you stop opening.

The promise in the first ninety seconds: The opening has to tell a viewer what question they will get answered and why it is worth the runtime. Draft the promise as its own card and test it cold, because everything downstream depends on it landing.

Sections you can see all at once: Lay out each section as a card with its point and its evidence, then look at the whole board. A soft middle is obvious in that view and invisible in a scrolling document, which is where most long essays quietly die.

Put the turn where it earns the ending: Every good essay has a moment where the argument complicates itself. Drag it earlier or later on the canvas and see what it does to the shape before you commit hours of narration to one arrangement.

Write it to be spoken: Draft narration beside the evidence it cites, then read it aloud and cut the clauses your mouth trips on. A script that reads well on the page and badly in the ear is the most common reason an essay sounds stiff.

A running source list, not a last-minute one: Because every card carries its link, the credits and the description list are assembled as you research instead of being rebuilt from browser history the night before you publish.

Share the board, get notes before the edit: Send a view-only link and a reader or an editor opens the structure in a browser with no account, so the argument gets challenged while it is still cheap to change.

Hand a plan to the edit, not a folder: Export the script and the beat-by-beat visual plan and take it into Premiere, Resolve, or Final Cut. Storyflow ends where the timeline begins, and it is deliberate about that boundary.