How to Lose the Plot
Start with hundreds of characters. Give each one a history, a set of relationships, a trajectory that shifts across four seasons. Add thousands of established facts—any one of which constrains every future script. Layer in temporal dependencies, thematic arcs, locations that carry the weight of everything that happened in them.
Now manage all of it in a spreadsheet.
That is how you lose the plot. Not metaphorically. Structurally. The complexity of what a writers’ room builds has outgrown the tools it uses to track what it’s built. Fabula is the context engine that closes the gap—a knowledge graph that reads your scripts, extracts everything in your show, and answers questions no document ever will. It updates itself. It cannot drift. And it already works.
Four Infrastructure Failures Nobody Talks About
These are not workflow problems. They are not memory problems. They are infrastructure failures—the predictable result of managing graph-structured data in flat documents. Every production encounters them. Almost none recognises what they actually are.
It’s Tuesday. Someone pitches a storyline and the room stalls. Not because the idea is bad, but because nobody can confirm whether you did something like it in Season 2. You look around. The showrunner is thinking. The script coordinator is scrolling through a spreadsheet that hasn’t been updated since before the hiatus. Nobody says “no,” but nobody says “yes” either. That silence costs you the momentum of the pitch and maybe the storyline itself. It is not a memory failure. It is a system that cannot answer “did we do this before?” across four seasons of television.
Fabula indexes every arc, character journey, and thematic thread across every season. The question “have we done this?” becomes a query that returns in milliseconds, with episode numbers attached.
A character from Season 1 is coming back. You need her title, her relationships, what she knew and when she knew it. You open the story bible. The story bible has a paragraph about her that was last edited eighteen months ago by someone who left the show. You open the scripts. There are eighty of them. The honest answer is that nobody remembers, and the current method for finding out is reading eighty scripts cover to cover. That is not a research method. It is an admission that no research method exists.
Every character has a canonical profile built automatically from the scripts—every appearance, every relationship, every piece of information they encountered, timestamped to the episode. Lookup takes seconds.
You’re reading tomorrow’s pages. A character mentions an event that contradicts something established two seasons ago. You catch it in prep. Good. If you hadn’t, it would have reached the stage, and the difference between catching it now and catching it in dailies is tens of thousands of dollars. But the real failure happened earlier: the system that stores your show’s canon has no mechanism to detect when new material violates it. Detection depends entirely on you remembering every established fact across every produced episode.
Fabula tracks every established fact, every relationship state, every piece of canon across your entire series. Contradictions surface before they reach the stage—because the system knows what has been established and can compare it against what is being written.
Legal needs a complete list of every real-world brand, location, and licensed music cue mentioned in this season’s scripts. By Monday. You sit down with a highlighter and start reading. Every script, cover to cover. You build the spreadsheet by hand. This happens every season. Sometimes it happens twice. The information was always in the scripts. The problem is that no system existed to extract it.
Every entity—object, location, brand mention, music cue—is extracted and cataloged automatically. Filterable asset inventories generated on demand. The Friday read becomes a Friday query.
Four Steps. One Graph. No Spreadsheets.
Fabula reads your scripts and builds a living knowledge graph of everything in your show. Not a summary. Not a spreadsheet. A queryable, connected, self-updating record of every entity and every relationship.
Upload Your Scripts
Final Draft, Fountain, PDF—standard industry formats, no special prep. Drop in a single episode or your entire series.
Fabula Builds the Graph
Characters, locations, relationships, objects, themes—extracted and connected automatically. Every entity gets a canonical profile. Every relationship gets a history. The graph updates when new scripts arrive.
Ask Questions Nobody Could Answer Before
“Who was in that scene in Season 1?” “Show me every character who shares scenes with the Captain and the First Officer.” “How did Worf’s loyalties shift after he joined DS9?” These are graph queries. They return in milliseconds.
Export What You Need
Story bible PDFs, asset inventories, character breakdowns—generated on demand, always current. The document is a view of the graph, not a separate artifact that someone has to maintain.
Lieutenant Worf. Worf, Son of Mogh. Ambassador Worf. One Person.
Fabula resolves aliases against a living canonical graph. “Lieutenant Worf,” “Worf, son of Mogh,” and “Ambassador Worf” collapse to a single canonical entity—one node carrying the full history of a character whose name changes with rank, with cultural context, with career. The system understands that a promotion doesn’t create a new character. A Klingon naming convention doesn’t create a new character. An ambassadorial appointment doesn’t create a new character. It understands this across seven seasons and 178 episodes, with zero drift.
That’s what entity resolution looks like at production scale. Not a string match. Not a similarity threshold. A system that understands narrative identity the way your script coordinator does—except it doesn’t leave the show in Season 3.
Your Data Leaves When You Leave
No proprietary formats. No vendor lock-in. Your show’s data is yours, in the formats your team already uses.
Excel and CSV
Asset lists, character breakdowns, location inventories—ready for production binders and legal review.
PDF Story Bibles
Formatted, print-ready story bibles generated from the live graph. Always current because they’re rendered from the graph, not maintained by hand.
API Access
Query your show’s data programmatically. Integrate with existing production tools, or build something we haven’t thought of yet.
The Infrastructure Your Stories Deserve Already Exists
We’re opening early access to production teams working on series television. Explore the live catalog and see what a knowledge graph does that no binder ever could. Or tell us about your show and we’ll tell you what Fabula can do with it.