The IP Flywheel: How Holywater Turns Book Platform Data Into Vertical Drama Commissioning Decisions

Most vertical drama commissioning decisions are made the way film and television commissioning decisions have always been made: a creative executive reads a proposal, feels something about it, and places a bet. The instinct is real. The track record of that instinct is mediocre. The majority of greenlit series underperform. The majority of commissioning budgets fund content that does not cover its acquisition cost.

Holywater's commissioning model is built on a different premise. Through My Passion, we're able to test hundreds of books each month and greenlight the most promising titles for live-action production. That sentence, from Holywater CEO Anatolii Kasianov, describes a commissioning process that uses reader behavior data from a book platform to predict which story premises will generate vertical drama paywall conversion before a single frame of production is committed.

The model is not about liking stories. It is about measuring how many people read past chapter three, how long they stay per session, how often they return to a story they started, and what completion rate the first ten chapters produce. These are behavioral signals. They predict commercial performance in the vertical drama context with more reliability than any creative executive's instinct, because they are measuring real audience behavior from real readers rather than projecting hypothetical viewer behavior from a script.

Holywater was founded in Kyiv in 2020 as a digital startup, originally building My Passion, a mobile application for books. The company's founders wanted to forge an IP incubator by which creators could pitch in work that would become tomorrow's hit content. The vertical drama insight came from that foundation, not from the video production industry. The resulting platform is the clearest available proof that the data-driven commissioning model that mobile gaming applies to content testing is viable in the vertical drama format.

The Pipeline: From Book to Pilot to Series

The Holywater ecosystem connects four products in a sequence that transforms reader behavior data into distributed vertical drama IP:

My Passion is the book platform. It is the number one independent digital book publishing platform among American and European companies. Creators publish serialized fiction on the platform, readers engage with it, and the platform captures detailed behavioral data about how readers interact with every title: reading sessions, chapter completion rates, return visit frequency, and the specific story moments where readers drop off or accelerate.

My Muse is the AI-powered pilot generation platform. When a My Passion title generates strong reader engagement data, it enters the My Muse pipeline, where an AI-native production process generates a pilot series in days rather than weeks. My Muse is a leading platform for vertical series produced with the support of generative AI. The pilot is the concept test: a complete short series that demonstrates the story's performance in video format before the full live-action production commitment is made.

My Drama is the vertical streaming platform. It is the number one vertical streaming platform among US and European companies, with 15 million subscribers and content in over 30 languages across 190 countries. My Drama is where the validated content is distributed. Series that have passed the reader engagement filter in My Passion and the pilot performance filter in My Muse are produced for My Drama's subscriber base with a higher commercial confidence than a cold-start original commission provides.

Freebits is the ad-supported companion layer, extending the audience reach beyond the subscriber base.

The pipeline's logic is sequential and data-driven at every stage. My Passion generates the audience behavior data. My Muse tests the video format viability. My Drama distributes the validated content. Each stage gates the next with performance data rather than creative judgment.

What the Book Data Actually Measures

The specific behavioral signals that My Passion generates, and that Holywater uses to make commissioning decisions, map directly onto the vertical drama format's commercial requirements in ways that most content businesses have not yet recognized.

Chapter completion rate. A book chapter on My Passion is approximately the same narrative unit as a vertical drama episode. A story where readers consistently complete each chapter and immediately begin the next is demonstrating the same behavioral pattern that episode completion rate and next-episode continuation rate measure in vertical drama. The reader who cannot stop reading is the viewer who cannot stop watching. The chapter cliffhanger that drives chapter completion is structurally equivalent to the episode button that drives episode continuation.

Session length and return frequency. A reader who spends 45 minutes per session on a My Passion title and returns to it within 24 hours is a reader in an active parasocial relationship with the story. That engagement pattern predicts the kind of viewer investment that produces paywall conversion in vertical drama. The reader has already demonstrated willingness to give the story extended engagement sessions and to prioritize returning to it over other available options. Both behaviors are prerequisites for paywall conversion.

Drop-off chapter analysis. The specific chapter where reader drop-off concentrates reveals structural problems in the story's middle section. A book that loses readers consistently at chapter 12 has a chapter 12 problem. In the vertical drama context, chapter 12 maps onto the episode 12 to 15 range: the post-paywall establishment block where the series needs to continue advancing forward motion to retain viewers who have already converted. A story that demonstrates consistent reader retention through its middle chapters, rather than front-loading engagement and dropping off, is a story whose structural arc is designed for sustained viewing investment rather than for initial hook performance alone.

Genre signal and emotional register. The readers who engage most deeply with specific My Passion titles are readers who have self-selected into a specific genre category and emotional register. The data tells Holywater not only which stories perform but which audience is responding to them. A billionaire romance title that generates strong reader engagement from women aged 25 to 45 has already demonstrated the genre-audience alignment that the vertical drama format's demographic targeting requires. The commissioning decision is confirmed by audience data rather than inferred from genre category alone.

The My Muse Concept Test Layer

The My Muse pilot layer is the element of Holywater's pipeline that most directly addresses the vertical drama industry's most expensive production problem: the greenlight decision made before any performance data exists.

We launched My Muse, an AI-generated streaming platform that allows us to create series in days and test pilots quickly before scaling them on My Drama. The days timeline is the commercial significance. A My Passion title that has generated strong reader engagement data enters the My Muse pipeline and emerges as a short pilot series within days. That pilot is distributed to a subset of My Drama's subscriber base. The pilot's hook rate, episode completion rate, and continuation behavior are measured before the full production budget is committed.

The concept test layer compresses the information gap between the commissioning decision and the commercial performance data from months to days. A traditional commissioning decision for a live-action vertical drama series at $150,000 to $300,000 generates commercial performance data 4 to 6 months after the greenlight. A My Muse pilot generates commercial performance data within days of the My Passion engagement data that triggered it.

The production economics of AI-native content make this concept test layer viable. Each drama is conceived as an IP seed: once the video run ends, the story moves to the company's original product, the e-novel app My Passion, and may return as a sequel if both funnels make money. The result is an integrated loop that keeps one title alive for months and reduces the need to feed the algorithm with ever-new content.

The inverse flow is equally important. A My Drama series that generates strong subscriber engagement can return to My Passion as a sequel story, generating reader behavior data that then feeds back into the commissioning process for a My Drama sequel series. The IP flywheel is genuinely bidirectional: book data informs video production decisions, and video performance data informs book publishing decisions.

What Spark Me Tenderly Reveals About the Model

Holywater's biggest hit, Spark Me Tenderly, has generated more than 7 billion social impressions and $20 million in revenue, outperforming the average US theatrical box office per film in 2025. That commercial outcome is the result of the flywheel model operating correctly: a story that demonstrated reader engagement signals on My Passion was developed into a vertical series for My Drama, and the series generated commercial performance that validates the pipeline's predictive accuracy.

The 7 billion social impressions figure reflects something specific about the flywheel model's commercial advantage beyond the production economics: a story that has an existing reader community from My Passion arrives at My Drama with a pre-built audience that the platform does not need to acquire from scratch through paid campaigns. The readers who loved Spark Me Tenderly on My Passion were the early adopters who generated the social amplification that multiplied the series' reach beyond the subscriber base. The flywheel's reader-to-viewer conversion is a user acquisition mechanism that operates outside the paid social advertising economy.

The Model's Structural Advantages Over Conventional Commissioning

The Holywater commissioning model has four structural advantages over the conventional vertical drama commissioning approach that every production company without a comparable data pipeline should understand.

Risk compression. A conventional commissioning decision allocates $150,000 to $300,000 against a creative judgment call. A Holywater commissioning decision allocates production budget against a My Passion reader engagement signal and a My Muse concept test result. The production budget commitment follows the performance data rather than preceding it. The financial risk of a wrong commissioning decision is not eliminated, but it is compressed to the concept test cost rather than the full production cost.

Premise validation that is specific to the format. The My Passion reader engagement data is not general market validation. It is validation from a mobile-first, serialized fiction audience that reads on the same phone they will use to watch the vertical drama adaptation. The reader who pays for premium chapters on My Passion is already demonstrating the willingness to pay for serialized mobile content that the vertical drama coin-unlock model requires. The audience overlap between My Passion readers and My Drama viewers is not incidental. It is the pipeline's commercial foundation.

The loop creates compounding IP value. A conventional production company produces a series, delivers it, receives a licensing fee, and starts the next production from zero. Holywater produces a series, delivers it, watches the IP return to My Passion as sequel fiction, generates more reader data, uses that data to commission the sequel series, and captures the sequel's commercial value from a position of demonstrated IP performance. Each cycle of the flywheel is less risky and more data-informed than the previous one. The IP compounds in value rather than resetting with each production.

Volume at the concept stage, capital at the validated stage. The My Muse platform generates concept test content at AI-native cost. The selection pressure is applied before live-action production capital is committed. This is the precise inverse of conventional production economics, where the capital is committed before the performance data exists. Holywater can test dozens of concepts per month at AI-native cost and allocate live-action production capital only to the concepts that pass the concept test threshold. The result is a content portfolio where the capital allocation is correlated with demonstrated performance potential rather than with creative instinct.

The COL Group Parallel

Holywater is not the only company operating a version of this flywheel. COL Group's international expansion is built on adapting content from its Chapters and Kiss interactive fiction platforms into vertical drama series for its global distribution network. COL Group's pipeline from content engagement data on its fiction platforms to commissioning decisions for its video platforms is structurally equivalent to Holywater's My Passion to My Drama pipeline.

COL Group's content pipeline starts on My Passion, our independent book publishing platform. We test hundreds of books to gather data from our audiences on which stories resonate with them. The top-performing stories then move to My Muse, which is our AI video platform where vertical series are produced with the support of generative AI.

That direct description of the data-driven commissioning model, applied at a company operating at a fundamentally different scale from Holywater, confirms that the flywheel model is not a startup innovation. It is a structural approach to content risk management that the most sophisticated players in the format are converging on independently.

What the Model Signals for Production Companies Without Book Platforms

Most production companies do not have a book platform generating reader behavior data to inform their commissioning decisions. The question the Holywater model raises for them is not how to replicate the flywheel exactly but how to build the equivalent data-driven validation layer into their commissioning process using available tools.

The closest available equivalent for a production company without a proprietary book platform is the concept test series: a 5 to 10-episode AI-native production that generates real platform performance data before the full production budget is committed. The concept test generates hook rate, episode completion rate, and continuation rate data from real viewers rather than predicting those metrics from a creative judgment. It is not as comprehensive as My Passion's reader engagement data pool, but it produces the same type of forward-looking behavioral signal in the vertical drama format's own performance context.

The combination of existing IP with proven audience engagement, a web novel with demonstrated readership, a social media story world with documented follower engagement, or a branded IP with quantified audience relationship, and a concept test series that validates the video format adaptation is the closest available approximation of Holywater's flywheel for production companies building without a proprietary data platform.

Double Tap Films, Pratilipi's microdrama studio, operates on a structurally similar logic: adapting stories that have already been validated by readership data on the Pratilipi platform rather than commissioning original concepts. The studio's production model is built around adapting pre-validated intellectual property, enabling the studio to create content at significantly lower costs while using audience engagement data rather than conventional commissioning instincts to make greenlight decisions.

The pattern is consistent across every company that has built a data-driven commissioning model in the vertical drama space: reader or viewer behavior data precedes the production capital commitment. The companies that have built proprietary data platforms have a structural advantage in the commissioning process that compounds with each production cycle. The companies that are building toward this model now are doing so before the advantage is insurmountable.

Axis AI Studios Perspective

The Holywater flywheel is the most commercially sophisticated commissioning model operating in the vertical drama market today. Its specific mechanics are not replicable by production companies that do not have a book platform generating comparable reader behavior data. Its underlying logic is replicable by any production company that builds the equivalent data-driven validation layer into its commissioning process.

The underlying logic is: behavior data from real audiences predicts vertical drama performance more reliably than creative judgment alone. The data can come from a book platform, a social media community, a concept test series, or any engagement data source that measures real audience behavior with the story premise before the full production budget is committed.

AI-native production is the mechanism that makes the concept test layer economically viable for production companies without Holywater's scale. At $15,000 to $30,000 for a concept test series, the cost of generating real performance data before committing a full production budget is proportionate to the risk it manages. The production company that treats the concept test as optional is managing its commissioning risk with creative judgment alone. The production company that treats the concept test as mandatory is building the data-driven commissioning discipline that the format's most sophisticated operators have already identified as the structural advantage.

For production companies who want to build vertical drama content with data-driven commissioning validation built into the development process, reach out at business@axisaistudios.com.

FAQ

Can Production Companies Without Book Platforms Replicate the Holywater Model?

Not exactly, but the model's core logic is replicable without a proprietary book platform. The core logic is: use real audience behavior data to predict vertical drama performance before committing full production capital. The data source can be reader engagement on an existing book platform, social media follower engagement with a story world, or concept test performance data from a short AI-native pilot series. The Holywater model's advantage is the scale and specificity of its My Passion data pool. The concept test series produces a less comprehensive but equally format-specific equivalent: real viewer behavior in the vertical drama context rather than real reader behavior in the book context.

What Reader Engagement Signals Best Predict Vertical Drama Paywall Conversion?

Chapter completion rate and return visit frequency are the strongest predictors, based on the structural parallel between chapter cliffhanger mechanics and episode button mechanics. A story where readers consistently complete each chapter and return within 24 hours is demonstrating the engagement pattern that vertical drama's paywall conversion requires. Drop-off chapter analysis reveals structural arc problems that translate directly to drop-off episode patterns in the vertical format. Genre-audience alignment data, which reader demographics are engaging most deeply, predicts the platform's user acquisition efficiency for the vertical drama adaptation.

Does the IP Flywheel Work for All Story Genres or Only Romance?

Whether the vertical drama insight beyond romance remains to be seen is the honest state of the evidence. Holywater's most commercially successful titles are romance and dark romance. The flywheel's predictive validity outside those genre categories has not been tested at comparable scale. The structural logic, reader engagement data predicts viewer engagement, should hold across genre categories. The practical challenge is that My Passion's catalog reflects the genres that mobile readers most actively seek, which skews toward romance and emotional drama. A production company testing the flywheel logic in thriller, mythology, or crime drama needs a data source from a community already engaged with those genres rather than applying romance reader behavior data to a different genre's commissioning decisions.

Further Reading

For the concept test series that production companies without book platforms can use to generate the equivalent behavioral validation data described in this post, the guide to how to test micro drama concepts before full production covers the full concept testing methodology, cost structure, and performance metrics.

For how the licensed IP vs original IP decision interacts with the data-validated commissioning model this post describes, the licensed IP vs original IP in microdrama cost guide covers the full cost and downstream value comparison between building from validated IP and developing original premises.

For the franchise architecture that the IP flywheel's bidirectional book-to-video-to-book loop enables at scale, the guide to building a microdrama franchise covers how a single strong character generates sequels, spin-offs, and IP value beyond the first series.