Prove the episode works
before you write a line.

The cold open or hook: Lay out the pre-title sequence as its own beat and ask what it buys you. In television it sets the question the episode answers. In a podcast or a video series it is the first thirty seconds, and it is the only part most of your audience is guaranteed to hear.

Four-act, five-act, or three-act: Broadcast hours usually run four or five acts because the breaks are commercial. Streaming and most podcasts run three, because the break is wherever attention might wander. Pick the shape and the board lays out the acts, or work without acts entirely and just sequence beats.

Act breaks you can interrogate: Put a card at every break and write down the question it leaves open. Breaks that land on a reversal, a reveal, or a decision hold an audience. Breaks that land on someone arriving somewhere do not, and seeing all of them in a row makes the weak one obvious.

The button and the carry-forward: Mark the button or tag that closes the episode on its tone, and mark the thread that carries into the next one. Episodes that end cleanly and leave nothing behind are the ones that stall a series, and a board shows you that at a glance.

One track per storyline: The A story carries the episode, the B story runs underneath it, and the C story or runner is the light thread that keeps returning. Give each its own horizontal track and place a card wherever it genuinely appears, not wherever you meant it to.

Find the story that vanishes: A track that goes empty for a third of the running time is the storyline nobody remembers by the end. It is nearly impossible to notice in a linear outline and completely obvious when the tracks are stacked.

Make the B story rhyme with the A story: The B story usually exists to say something about the A story from a different angle. Put the two tracks next to each other and you can check that they are actually about the same thing, which is the difference between a subplot and filler.

Stagger the resolutions: When every track resolves on the same beat the ending feels mechanical. Seeing the resolution points lined up vertically is the fastest way to catch it and move one of them.

A first outline against your structure: Give it the premise, the format, the running time, and the structure and it lays out the acts with beat cards inside them. Starting from a shape you disagree with is far faster than starting from a blank template.

Ask it where the structure is soft: Ask what your act two break leaves unanswered, where the B story goes quiet, or what could happen between two beats that currently do not connect. It reads the whole board, so the answer is about your episode rather than a generic one.

It structures, it does not ghostwrite: Storyflow is a planning canvas rather than a script generator. The AI helps you see and arrange the episode, which is the genuinely hard part, not produce dialogue you did not write.

Re-prompt without losing work: Ask for a five-act version of the outline you already have and the cards you wrote and placed stay exactly where they are while the AI rebuilds the structure around them.

View-only links: Send a link and a showrunner, producer, co-host, or editor opens the whole outline in a browser with no account. Notes on structure arrive much earlier, which is when they are cheap to act on.

Outline together: A writers room, a co-writer, or a podcast duo can work on the same canvas, so one episode has one outline rather than two versions in two inboxes. Anyone a paid member invites to a board joins free today.

Keep characters and reference on the board: Pin character profiles beside the arcs they drive, drop in research, transcripts, images, and links, and grab frames from YouTube or Vimeo. The detail the episode depends on sits next to the beat that needs it.

Export for the notes call: Export the outline board as a high-quality image or PDF when it has to travel into a deck, a network notes call, or a document someone wants to mark up.