Every method on this page attacks the same problem: a screenplay is a sequence machine, and drafting it top-to-bottom is the worst way to discover its sequence. The tools below all pull structure out in front of prose — so the expensive material (scenes, dialogue) gets written against a shape that already works.
The 70-card method
David Lynch, who learned it from Frank Daniel at the American Film Institute.
Take seventy 3×5″ index cards. Each card gets a scene — its name or header at the top, then whatever fleshes it out: keywords, a line of dialogue, who's in it, how it opens, how it ends. Fill all seventy before writing anything that looks like a screenplay. Only then sit down and write the scenes so they fit together narratively.
The point of the cards is physical: you can resequence the film on the table, ahead of time, when a move costs nothing. Seventy scenes at roughly a scene per page-and-a-half is a feature — the count isn't mystical, it's a budget that forces the whole film into view before any one part of it gets precious.
Frank Daniel is also the source of the "sequence approach" — the feature as eight self-contained sequences — which is the same instinct at a coarser grain.
But / therefore
Trey Parker & Matt Stone, from their much-circulated storyboarding lecture.
Read your beat list aloud. If consecutive beats connect with "and then," the story is a list, not a story. Every join should be a but or a therefore: this happens, but that blocks it; therefore she does this. Causality is the whole test — each beat must be caused by the one before it or collide with it.
"And then" is where a script dies. "But" and "therefore" are where it moves.
It's the cheapest structural diagnostic there is: no software, no cards, just conjunctions. The Breaking Bad writers' room worked the same muscle at series scale — every board beat earned by the one before it.
Want, plan, failure
David Mamet — the drum he beats hardest in On Directing Film.
In every sequence, ask three questions: what does the character want right now, why don't they get it, and what do they do about it? A want produces a plan; the plan fails; the failure produces a new plan. That loop is the scene, and chaining the loops is the plot.
The discipline this buys is simplification. You never have to invent "what happens next" from nothing — you only have to know what the character wants now, and put the obstacle in honestly. Intent drives action; action meets resistance; resistance forces the next intent. Works on cards, a whiteboard, a notebook — the substrate doesn't matter, the question does.
Beats in brackets, no dialogue
Working practice from the same thread this page distills — a drafting tactic, not a famous name's.
Blast the whole story down first with no dialogue at all — just short bracketed beat-text, sequences split by white space. It reads less like a screenplay than like a shot list of intentions. The rule is speed: get every sequence onto the page while the shape is still cheap to change.
Only when the bracketed skeleton tells the story end to end do you open a sequence up and write actual scenes. Dialogue written before the shape settles is dialogue you'll defend instead of delete.
Start where it's exciting — and hold act three loosely
Same source — and the most honest note in the thread.
When the skeleton is done, don't write from page one. Go to the most exciting sequence in act one or two and write that — never act three. The reasoning is blunt: by the time you're halfway through act two, you finally know the characters, and everything in the back half changes anyway. Act three as first-drafted was written by someone who hadn't met them yet.
So don't fall in love with the late pages. They're scaffolding by an earlier, more ignorant version of you — budget your affection for the material that will survive.
Provenance
Distilled from a Reddit thread on Lynch's card method — the 70-card account via commenter uglyorgans, the drafting practices via commenter PencilWielder — cleaned up and cross-checked against the named sources (Lynch on Frank Daniel's AFI teaching; the Parker/Stone lecture; Mamet's On Directing Film). Synthesis and errors mine.