Twenty-Three Episodes Later, You Still Can't Name the Actors

You open your phone to check one thing. Forty minutes later you have watched twenty-three episodes of a show about a housekeeper who is secretly a billionaire's estranged wife, you have paid $4.99 to unlock the next batch, and you could not name a single actor in it.

If that sounds absurd, consider that it is now one of the fastest-growing businesses in entertainment. Shows shot vertically, cut into sixty- to ninety-second episodes, and sold by the unlock have gone from a curiosity to a category that analysts size in the billions — and that Hollywood studios and European broadcasters are quietly staffing up for.

Here is what this format actually is, why it grew so fast, and what it tells us about where screen storytelling is heading.

What a vertical drama actually is

The format goes by several names — vertical drama, microdrama, short drama, or duanju, the Chinese term where it started. The mechanics are consistent across all of them.

A series is shot in 9:16 portrait orientation, framed for a phone held in one hand. Episodes run roughly 60 to 120 seconds. A full season is not eight episodes but 60 to 100 of them, which means a complete story arc lasts about as long as a feature film — just delivered in a hundred pieces instead of one.

The story structure is where it gets interesting. Because the viewer can swipe away at any second, every episode is built to end on a hook. Not a chapter-ending hook of the kind a prestige drama uses at the end of an hour, but a hook every ninety seconds, relentlessly. A slap lands. A paternity test arrives. Someone walks into a room they should not be in. Cut to the paywall.

The unit of storytelling is no longer the episode. It is the cliffhanger.

Production matches the economics. A vertical series is typically shot in one to two weeks with a small cast, a handful of practical locations, and budgets that industry reporting generally places in the low hundreds of thousands of dollars — a fraction of what a single streaming episode costs. That speed is the whole point: a studio can test ten premises in the time a traditional show takes to finish a pilot.

How big this got, and how fast

The numbers are genuinely hard to pin down, and you should treat any single figure with suspicion — different analysts count different things, and some count mainland China while others exclude it. But the direction is unambiguous across every source.

Two apps dominate the Western-facing side of this: ReelShort and DramaBox. Both crossed 50 million monthly active users during 2025, and by most estimates the two together account for a large majority of short-drama in-app purchase revenue outside China. Media Partners Asia has put ReelShort near 29% of ex-China microdrama revenue with DramaBox around 21%.

The spread between "$3.6 billion" and "$11 billion" is not sloppiness on anyone's part. It reflects a real question about what you are measuring — app-store purchases only, or advertising and licensing too; ex-China or global. What matters for the reader is the shape: a category that was a rounding error in 2023 and is now large enough that Fox Entertainment Studios has committed to producing hundreds of vertical titles, and that UK broadcasters are actively investing in.

Why it works, mechanically

It is tempting to write this off as attention-span decay. That explanation is too easy, and it misses three structural things the format gets right.

It monetizes at the moment of maximum desire. A subscription asks you to pay before you know whether you will like anything. A vertical drama asks you to pay at the exact instant you need to know what happens next — usually around episode 10 to 20, after you are already invested. The price per unlock is small, often a dollar or two in coins, so the decision feels trivial. The cumulative spend across a season is not trivial at all, which is precisely why the model works.

It removes the choice problem. The hardest part of a streaming subscription is the fifteen minutes you spend scrolling a grid deciding what to watch. Vertical apps open into an autoplaying feed. There is no decision to make; the first episode is already running. In a market where every platform is competing for the same evening, eliminating the decision is a real product advantage.

It fits the gaps in a day. Ninety seconds fits a bus stop, a queue, a lunch break, the four minutes before a meeting starts. Traditional television competes for a viewer's evening. This format competes for the minutes nobody was using — which is a much less crowded market.

The genres that actually perform

The catalogue is narrower than you might expect, and deliberately so. Romance is the volume driver: secret-billionaire plots, contract marriages, revenge arcs, and the "underestimated protagonist who turns out to be powerful" premise recur endlessly because they reliably convert browsers into first-time payers.

Thrillers and revenge dramas play a different role. They are generally weaker at top-of-funnel reach but stronger at retention — viewers who start one are more likely to finish the full arc. The studios doing best in 2026 reportedly run both deliberately: romance to acquire the audience, thriller to keep it.

Romance brings them in. Thriller keeps them watching.

Werewolves, secret heirs, and time-loop revenge remain stubbornly popular. The genres that struggle are the ones that need room to breathe — slow character studies, ensemble ambiguity, anything where the payoff is at the end rather than every ninety seconds. That is not a flaw in the audience. It is a constraint of the container.

What this means for people who make things

If you write, act, direct, or edit, this category is worth understanding even if you never work in it, because it is changing three things at once.

It is creating a new entry-level tier. Vertical productions shoot fast, hire constantly, and are far more willing to take unknown performers and first-time directors than a prestige series is. For a lot of people this is becoming what regional theatre and daytime soaps once were — a place to accumulate credits and reps.

It is also compressing the feedback loop. A studio can greenlight a premise, shoot it in ten days, and know from retention curves within a week whether the concept works. Traditional development takes eighteen months to learn the same thing, and usually learns it from a focus group rather than from paying customers.

And it is teaching an uncomfortable lesson about pacing. The argument that vertical dramas are simply badly made misses that they are optimized for a metric traditional television never had to face: whether a specific viewer swipes away at second 40. That is a brutal, honest measurement. Some of what it rewards is genuinely cheap. Some of it is craft — a well-placed reveal, a clean edit, an actor who can land a turn in three seconds.

What to watch for next

A few tensions are worth tracking if you follow this space.

Quality escalation versus unit economics. As Hollywood money enters, budgets rise. But the whole model depends on a title being cheap enough that ten failures are affordable. Nobody has yet shown that a prestige-budget vertical series earns back the difference.

Advertising versus unlocks. The pay-per-unlock model extracts a lot from a small number of heavy spenders. Several platforms are testing ad-supported free tiers, which is a completely different business with completely different content incentives.

Market saturation. Download growth cannot compound at these rates forever. When acquisition slows, the competitive question shifts from "who can make the most titles" to "who can make titles people finish" — and that favors people who can actually write.

Format bleed. The more interesting possibility is that vertical techniques migrate outward. Faster cold opens, tighter act breaks, more aggressive hooks in conventional streaming shows. If that happens, this category will have changed television without most viewers noticing.

The short version

Vertical drama is serialized fiction shot for a phone: 9:16, one to two minutes an episode, 60 to 100 episodes a season, paid for by the unlock rather than the month. It grew from near-nothing to a multibillion-dollar category in roughly three years by solving three problems at once — charging at peak desire, removing the choice paralysis of a content grid, and fitting into the minutes between other things.

Whether the storytelling is good is a fair debate and largely a matter of taste. Whether it matters is not really up for debate anymore. Figures cited here are drawn from industry analyst estimates as of this writing and vary considerably by source and methodology — the trajectory is far more reliable than any single number.

If you want to understand it, the fastest way is to give it ninety seconds and watch what it does to you. That is, after all, exactly what it was built for.