Overview
Built a three-part AI suite that finds adaptable story material, maps projects onto a 139-entity story ontology and diagnoses scripts on an 11-dimension scorecard, designed to advise development teams rather than write for them.
Who This Is For
TV production companies, development teams and story consultants who assess and develop scripted drama. This case is most useful if:
- script assessment depends on whoever happens to read the draft
- you want story frameworks applied consistently across a slate
- you need to compare a project against the genre it belongs to
- you want AI to diagnose and advise, not to write
The Challenge
The studio's development team assessed treatments, outlines and scripts against a wide range of story frameworks, but each reading relied on individual expertise and memory.
It needed a repeatable way to diagnose story structure, compare projects with genre baselines and find promising new material, without turning the creative work over to a machine.
Our Approach
We built a three-part AI story intelligence suite, explicitly designed as a diagnostic and development system rather than an AI screenwriter.
- Story Radar (finding): scores real articles for adaptation potential on a 10-dimension, 100-point scorecard with a pass, archive, watchlist or develop recommendation, generates premises from a story taxonomy, and modernises a verified library of 107 pre-1920 "forgotten plots".
- Narrative intelligence platform (knowing): a story ontology of 139 entities across 11 domains, including 40 masterplots, 22 tropes and 15 themes, cross-referenced to classic plot typologies, with 37 TV subgenres and full baselines for four of them.
- Narrative OS (diagnosing and developing): reads scripts and treatments in common formats including Final Draft and Fountain, and scores them on an 11-dimension weighted scorecard and a 5-level narrative maturity model.
- character work through Enneagram types, Jungian archetypes and dramatic functions
- executive coverage, drift audits, season arcs, beat sheets and scene outlines
- AI throughout: Claude models handle analysis and categorisation, including an agent that maps a project's outline onto the ontology, with model tiers chosen per task to control cost.
Impact
Consistent diagnosis across a slate. Every project is assessed on the same 11 dimensions and maturity scale, so development notes compare like with like.
Genre context on demand. Projects can be set against TV subgenre baselines rather than one reader's sense of the genre.
Verified, not invented. The forgotten-plot library was checked against full source texts: 104 of 107 records passed, after verification caught errors such as wrong publication years.
AI is good at applying frameworks tirelessly and bad at taste. This suite gives the first job to the model and keeps the second with the development team.
Built and delivered by ECM.DEV. Client details are withheld at the client's request.
Topics
Delivered Norwegian-to-English translation and QA review of SpareBank 1 Markets' website content, covering their research, corporate finance and capital markets service lines.
Built a data-driven content and channel strategy for Econa, Norway's professional association for economists and business graduates, building on an already-delivered website redesign to reduce member churn.
Built a CRM-connected advertising sales platform for a leading Nordic digital marketplace. Connected audience data, targeting options, advertising formats, pricing and sales contacts into a single self-service portal for advertisers and media agencies, replacing a fragmented sales experience with a repeatable path from research to qualified inquiry.
The Content Lifecycle Redesigned
How AI Changes Every Stage from Creation to Retirement
What agents need to operate your content
The article discusses the infrastructure and capabilities required for AI agents to effectively manage content operations in enterprise environments, focusing on the practical requirements developers must understand when building systems that automate content workflows.
Content-to-Pipeline Check
See how far your content connects to sales, and how much of what sales learns comes back.