We modelled a national launch month by month. Where the money actually goes, why the catalogue is almost free to hold, and why the cost to serve each viewer drops as you grow.
The question came, as it always does, from the person writing the cheque. Nina is building a microdrama app: ninety-second vertical episodes, cut on cliffhangers, a few free before a paywall. She had the content plan, the acquisition plan, and the pitch. What she did not have, until an investor asked it plainly, was an answer to the only question that decides whether any of it works. When a viewer watches thirty episodes back to back on their phone, and then another thirty tomorrow, does the cost of streaming all that video eat the dollar you make from them?
It is a fair fear. Microdrama is a binge machine, and binge machines move a lot of minutes. So she did what you do when a number decides everything. She built the model, month by month, and let the arithmetic answer.
What a microdrama platform actually is
Strip away the storytelling and a microdrama app is a volume-of-minutes business. Episodes are short, ninety seconds each, but there are hundreds of them and viewers watch a lot in a sitting. The catalogue grows fast, dozens of new episodes a month. And the audience, measured in monthly active users (MAU) and the daily active users (DAU) who actually open the app, climbs if the content lands. Every one of those viewers streams video in multiple resolutions, the adaptive bitrate (ABR) renditions that keep playback smooth on a train or on wifi. The bill is driven by one thing above all others: minutes streamed.
The model runs a realistic national launch: 5,000 monthly actives in month one growing to about 14,000 by month twelve, a catalogue climbing from 50 episodes to over 1,000, and each daily active viewer watching around fifteen minutes. With daily actives at roughly a third of the monthly audience, that is about 675,000 minutes streamed in month one alone, on the way up.
Where the money goes
Here is the first surprise. A full year of running that platform costs about $32,000, and almost all of it is two line items.
Streaming is 46 percent of the bill and the fixed platform fee, which covers the player and infrastructure across Android, iOS and web, is another 45 percent. Professional services for integration make up most of the rest. Encoding and storing every episode you own, all thousand of them by year end, comes to about $100 for the entire year. Read that again: the content itself, the thing the whole business is about, is a rounding error on the invoice. Short episodes are tiny to encode and tiny to store. What you actually pay for is delivering them and running the apps that play them.
The number that decides it
Total cost is the wrong lens for a subscription business. The right one is cost per viewer, and this is where the model answers the investor.
In month one it costs about 43 cents to serve an active viewer. By month twelve it is about 24 cents. It falls the whole way, and the reason is structural: the fixed platform fee spreads across a bigger audience every month while per-minute delivery stays flat, so the unit cost keeps dropping as you scale. Against an assumed dollar of revenue per active viewer, the gap is your margin, and the margin widens as you grow rather than narrowing. That is the shape of unit economics a founder wants to see, and it is the opposite of the fear that streaming costs would swallow the business.
A binge business is supposed to punish you for engagement. On per-minute infrastructure, more watching means more revenue against a unit cost that is already falling.
Adding episodes is nearly free
Microdrama lives or dies on content velocity, the relentless drip of new series that keeps viewers opening the app. The good news from the model is that velocity barely touches the bill.
Encoding fifty new ninety-second episodes costs about three dollars. Holding the entire catalogue, more than a thousand episodes at that year-end peak, costs under fifteen dollars a month. Adding a series is not a budget meeting; it is a publish button. Experiments that used to be capital decisions become playlist decisions, which is exactly the freedom a content-hungry format needs.
Why the bill is predictable
Because everything is priced per minute, the cost tracks actual watching. There is no bundled allowance to breach and no overage cliff waiting in a month when a series goes big. The model lands at a month-twelve run rate near $3,400 a month, and every line of it is knowable in advance. You can price your own launch on the pricing page before you commit to anything, then change every input in the downloadable model to match your content plan and your growth curve.
What you are actually buying
The line items above buy more than delivery. The pieces a microdrama app needs run on one API and one per-minute bill.
| Capability | On FastPix |
|---|---|
| Video on Demand + Player | Upload an episode, get a playback URL. A vertical, mobile-first player on iOS, Android and web, your brand and your controls. |
| In-Video AI: clips + reframes | Turn an episode into vertical clips and teasers for TikTok and Reels automatically, captions burned in because social video is watched on mute. This is your acquisition engine. |
| Video Data | See the episode funnel and paywall conversion: which episode hooks, where viewers drop, which one to put the wall after. 56 dimensions, ten yours to define. |
| Content protection | Signed playback (JWT), domain lock, and DRM for paywalled episodes when you need it. Excluded from this model’s figures. |
The growth loop this pays for
The economics only matter because of the loop they fund. A microdrama platform grows by cutting clips from its own episodes and posting them relentlessly where discovery happens, then turning the installs those clips drive into binges, then walling the binge at the cliffhanger. In-Video AI cuts and reframes the clips, the Player holds the binge, and Video Data tells you where the wall converts and where it just loses people. The same per-minute infrastructure that keeps the bill low is also the machine that fills the top of the funnel.
What FastPix does not do
The model is honest about its edges. FastPix is not your paywall, your coin economy, or your billing system, and it is not your recommendation engine. You bring the monetization and the app; we are the video infrastructure underneath them. The figures here also exclude DRM and payment processing, both of which you may add. If what you need is a finished consumer app rather than the video engine to build one on, that is a different tool, and we would rather say so here than later.
Nina got her answer, and it was the one the format needed: the content is nearly free to hold, the cost to serve a viewer falls as the audience grows, and the video bill is a line she can forecast rather than fear. Ship the pilot. Cut the clips. Watch the funnel. Price the catalogue. Modern video is what we do.
Frequently Asked Questions (FAQs)
How much does it cost to run a microdrama app?
In a modeled national launch growing from 5,000 to about 14,000 monthly active users over a year, total video infrastructure cost is around $32,000 for the year, or approximately $0.24 to $0.43 per active viewer per month depending on scale. Streaming and a fixed platform fee are the two largest cost components, while encoding and storing the entire catalog costs less than $100 for the year. Adjust the downloadable model with your own usage assumptions.
What drives the cost of a microdrama platform?
Minutes streamed are the primary cost driver. Because episodes are short but heavily binged, video delivery is the largest variable expense alongside a fixed platform fee. Encoding and storage contribute very little because 90-second episodes are small, while professional services and integration costs are more significant during the initial setup.
Do the unit economics of microdrama work?
In this model, yes. Cost per active viewer decreases from about $0.43 in the first month to about $0.24 by month twelve as the fixed platform fee is spread across a larger audience, while delivery costs remain proportional to viewing time. Assuming approximately $1 of revenue per active viewer, margins improve as the platform grows. Replace the revenue assumption with your own business metrics for accurate projections.
How much does it cost to stream a 90-second episode?
Using the Premium 1080p pricing model described here, encoding a 90-second episode is a one-time cost of only a few cents, while delivery is billed per minute streamed. Since each episode is short, storage and encoding costs remain minimal, making total viewing time the primary factor affecting streaming costs.
What is per-minute video pricing?
Per-minute video pricing charges separately for minutes stored, minutes delivered, and one-time encoding during upload. Unlike plans with bundled usage allowances, this pricing model aligns costs directly with actual video usage, making expenses easier to predict as your platform grows.
