How to Write a Vertical Micro Drama Screenplay

October 23, 2026 • By Greenlight Writing

A vertical micro drama is a scripted serial made for a phone held upright: episodes of about 60 to 120 seconds in a 9:16 frame, dozens of episodes per season, each one opening on conflict and ending on a cliffhanger. The format goes by several names (vertical drama, short drama, mini drama, duanju), but they describe the same thing: a story told in one-minute pieces that viewers binge by tapping next. The trade press describes a typical series as a 60 to 100 minute story divided into roughly 90-second episodes, each ending with a cliffhanger.

To write one, plan the whole season first, then break it into one-scene episodes of roughly 180 to 360 spoken words. Each episode must put its conflict on screen inside the first five seconds, turn at least once near the middle and end on a hook that makes the next episode feel non-optional. That is the whole job. Ninety times in a row, if your season runs to ninety episodes.

The episode is the unit, not the act

A feature screenplay is organized by acts. A vertical micro drama is organized by episodes. Final Draft puts the typical season at 60 to 80 episodes of one to three minutes each, shot in about a week. The apps in the ReelShort and DramaBox mould have settled on a visible norm of 60 to 100 episodes of 60 to 120 seconds.

This is the mistake almost every newcomer makes: writing a 100-minute feature and chopping it into hundred one-minute pieces. Final Draft states it plainly. You cannot do that. Each micro episode must contain its own beginning, middle and end, along with its own cliffhanging beat to keep viewers binging along. Every episode needs its own building conflict, emotional stakes, a small piece of character arc and a compelling cliffhanger, all inside a few script pages at most.

Think of the season as a feature film with the act breaks sawn off and replaced by cliffhangers. A three-act spine still sits underneath, but nobody watches the spine. The audience watches the next sixty seconds, and the writing job is to make the next sixty seconds non-optional, episode after episode.

The episode anatomy: hook, turn, cliffhanger

Episodes this short have no room for setup. They open mid-conflict or on the previous episode's twist, because viewers arrive by swiping and leave by swiping. The practical guide from SceneMixer, checked against the major apps' 2026 catalogues, lays out three beats every episode must hit:

• Conflict inside the first 5 seconds. The first line or the first action is the conflict itself, not the setting. A question like "Did you burn the contract?" beats an establishing description of the house it was burned in.
• A turn about every 30 seconds. A 90-second episode turns at least twice: new information, a shifted allegiance or escalated danger. The viewer should never be able to coast.
• The episode-end hook, rotating. This is the beat that buys the next tap, and it comes in three types (see the next section). The episode cuts on the question, one beat before the answer.

JellyPod's episode diagram describes the same shape in four words: a hook, an escalation, a reveal and a cliffhanger across about one to two minutes. Call it beats or call it anatomy, the discipline is identical. Open fast, move the conflict, cut before resolution.

Three hook types, and why they must rotate

SceneMixer's format guide names three hook types that carry the end of every episode:

• Identity. Someone is not who they said they are.
• Emotion. A confession, a refusal or a rupture, left hanging.
• Realisation. The viewer and the lead discover together that an earlier judgement was wrong.

Rotate them. Two identical hooks in a row feel like a trick, and the audience learns to discount the pattern instead of feeling it. Label each episode's hook with its type when you outline, so you can check the rotation yourself before you write dialogue. If you cannot describe an episode's tension in one line, it is not ready to script.

The first three episodes decide everything

Retention decides this format, so the opening of the season is engineered, not inspired. The first-three-episodes beat sheet from SceneMixer's guide works as a template:

• Episode 1: conflict inside five seconds, the lead's situation and first trap, and what she wants made clear. End on an identity hook.
• Episode 2: the antagonist enters, the stakes go on the table (money, a marriage, a life, a reputation: pick one), and the lead is cornered. End on an emotion hook.
• Episode 3: the first reversal. The apparent weakling shows a card, or an ally turns. End on a realisation hook.

The logic is commercial and should shape the outline from day one. Free-to-watch feeds decide on completion of the first three episodes, and the paid apps convert at the paywall around episodes 8 to 12, where TheWrap reports users get the first eight to 10 episodes free and then must pay to unlock the rest. The lead's first real win must land by episode 5 or completion collapses before the paywall ever appears. Put your strongest twist around the point where the free episodes run out, since that is the moment viewers decide whether to pay. In this format, commerce is structure.

Write for the phone, not the page

A phone caption line holds about six to eight words. A long dialogue line wraps to three rows and the viewer ends up reading captions instead of watching the actor's face. So dialogue must be extremely lean: SceneMixer recommends keeping single lines under about 20 words, and JellyPod's production advice adds the filming rules that change how you write: frame for 9:16 from the start rather than cropping a horizontal shot, favor close-ups, keep faces clear of the caption area and the app buttons at the bottom of the screen.

A few more habits from the same sources, adapted for an independent writer:

• Keep one spelling of every location for the whole season. Consistency is what a shoot day runs on, whether the crew is twenty people or three friends.
• Mark the hook at the end of every episode in your outline, with its type, so the rotation is checkable.
• Convert dialogue to time as you draft. At a typical spoken pace of about three words per second, a 90-second episode is roughly 270 words of dialogue and voice-over; action lines are not counted because they are not spoken. Work the formula backwards after every draft instead of guessing the running time.
• Cut dead air in the edit. Burn in captions for viewers watching with the sound off. Check that the first frame of each episode makes sense on its own, because that is the frame the thumb decides on.

There is no single mandated template for the major apps. Producers and studios generally accept a standard screenplay layout organized by episode: a scene heading with INT. or EXT., location and time, action lines, character name and dialogue, and a labelled hook at the end of each episode. If you want to learn that layout properly first, the guide to formatting a screenplay covers the mechanics this format still borrows. Clarity and pacing matter more than any house style.

What most format guides leave out

The format guides ranking today are written for a specific reader: a writer submitting to the big apps. They leave out the parts that matter most to independent writers making their own series:

The production reality is industrial. TheWrap reports that a vertical series typically costs $150,000 to $300,000 over an eight to 10 day shoot, and that ReelShort's US team of 20 writers turns out 15 to 20 scripts a month. You are not competing against casual creators. You are competing against a pipeline. That does not mean you cannot play. Final Draft notes that reported budgets run $50,000 to $250,000 for seasons of limited locations and sets, and that a newcomer with friends, peers and a smartphone can shoot for far less. The format's speed is your opening: batch every scene at one location into the same shooting day, keep the principal cast small so viewers recognize each face at a glance, and let the cliffhangers do the work that production value does in a feature.

The genres are narrow, for a reason. Micro dramas rely heavily on familiar, melodramatic tropes: high drama, secret romances, revenge arcs, the compacted equivalent of modern soap operas. Billionaire and CEO romances are a staple, and ReelShort's catalogue is mostly romance with thrillers now entering. Final Draft notes that horror, sci-fi, action and comedy are less prevalent, primarily because genre production costs more. If your series lives in one of those genres, the cliffhanger engine still works, but budget the spectacle honestly before you write it.

The free run is a sales funnel. If you plan to release through the apps, the first eight to 10 episodes are free and the buy decision happens the moment the paywall drops. The free run must therefore climax on the biggest hook of the entire series, not warm up toward it. Every downstream choice, where reveals land, how arcs chunk, when the recognition scene detonates, is made with the retention curve and the paywall in view.

The market in one paragraph

The format began in China as duanju, short dramas that appeared around 2018 on Douyin and Kuaishou and spread to WeChat's video platform. By 2024 the domestic market was worth 50.44 billion yuan, larger than China's 42.5 billion yuan box office that year. The apps then crossed over: ReelShort launched in August 2022, DramaBox in April 2023, and in Western markets reported revenue surpassed $1 billion in 2024, with projections of more than $14 billion globally by 2027. Sensor Tower's State of Mobile 2026 report puts short drama apps at $2.98 billion in in-app revenue worldwide in 2025, and Deloitte predicts that figure more than doubles in 2026. A young format with real money behind it is exactly when an independent writer should learn the craft, while the rules are still being written.

Keep the story yours

The cliffhanger engine is a structure, not a story. It cannot tell you what your lead wants, what she fears, or what the final hook costs her. Those choices are the screenplay, and they belong to the writer alone. Use any tool, including AI, the way you would use a good teacher: to ask questions about work you have already written, to study structure, to check your format. GREENLIGHT's AI Policy states the line plainly: AI cannot write your screenplay, invent your characters or decide what your story should be. The tools stay in their supporting role, and the story originates with the creator.

If you are learning the craft from the ground up, the GREENLIGHT Academy teaches structure the same way this article does, one decision at a time, so you can apply it yourself on the Create page. And because the work is yours, read the Protect Your Work page before the series leaves your desk. Writers developing the same instincts at feature length should also read the guide to writing a short film you can actually shoot, which applies the same shootable-thinking discipline at a different scale.

Frequently asked questions

How many words is one episode of a vertical micro drama?

About 180 to 360 words of dialogue and voice-over. At a typical spoken pace of three words per second that is 60 to 120 seconds of screen time. Action lines are not counted because they are not spoken. A 60-episode season is roughly 10,800 to 21,600 words.

Do the apps publish a standard script format?

No. The major apps do not publish a single mandated template. Producers and studios generally accept a standard screenplay layout organized by episode, with a scene heading, action lines, character name and dialogue, and a labelled hook at the end of each episode. Keep dialogue lines under about 20 words and use one spelling per location and character name across the whole season.

Can one person make a micro drama?

Yes. A solo creator can shoot a small live-action series on a phone with a few actors, keeping the cast small, locations limited and captions burned in. The writing discipline matters more than the crew size: one scene per episode, conflict in the first seconds, a turn in the middle and a cliffhanger at the end.

What genres work best?

Romance, revenge, hidden identities and second-chance stories dominate the English-language catalogues, because the cliffhanger engine runs on heightened emotion and moral clarity. Genre material like horror, sci-fi and action works too, but costs more to produce, so budget the spectacle before you write it.

Sources

• What Are Verticals and Micro-Dramas?, Final Draft
• What Is a Micro Drama? Vertical Dramas Explained, JellyPod
• Micro Drama Script Format, Template and Example (2026), SceneMixer
• AI Policy, GREENLIGHT
• All sources were read on October 23, 2026. Market and production figures above follow those sources; trade-press figures are attributed to the outlets those sources cite.