Skip to content
SparkFire
Case study · Media & Publishing

Manuscript Intelligence for Publishers

AI-assisted acquisitions and editorial intelligence system for a major trade publisher — surfacing comp titles, market signals, and editorial flags across hundreds of incoming manuscripts a week.

Generative AIKnowledge SystemsCustom Software
12×

Faster acquisitions screening

200M+

Tokens processed daily

8

Imprints onboarded

Challenge

Editorial teams were spending hours per manuscript on first-pass screening — reading samples, hunting for comps, judging market fit — while the inbound queue kept growing. Institutional knowledge lived in a few senior editors' heads.

Approach

  • Custom knowledge graph linking manuscripts to comp titles, sales history, and reviewer notes from the publisher's archive.
  • LLM-powered manuscript synthesis that produces editorial briefs with citations back to the source text.
  • Reviewer workspace built around the editor's actual workflow — not a generic AI chat.
  • Production-grade retrieval, 200M+ token-per-day inference budget, and per-imprint cost dashboards.

Outcome

Acquisitions screening now runs 12× faster. Eight imprints are live on the platform, processing over 200M tokens daily, with editorial decisions documented and auditable from inbound to offer.

Got an idea? Let's make it real.

Tell us about your problem. We'll come back within one business day with a take, a rough plan, and a call invite.