What we can build on top of your own wire

Six capability tracks, all already proven against a live 6.4M-article Benzinga corpus — not a proposal built from slideware.

Any chatbot can fetch today's headlines. Almost none of them can tell you which headline actually moved the stock.

Retrieval is the one thing LLMs already do better than any wire product — a frontier model with web search pulls recent news on any ticker for free. That makes raw "we have the news" a shrinking asset. The durable value sits one layer up: things that require holding the whole corpus at once and measuring it against what actually happened next. That's the layer described below, and it's already running.

Running today, on your feed

This isn't a pitch deck — it's a status report

6.4M
Articles ingested, 2016–present
126M
Sentences embedded (e5, 768-dim)
14
Vector collections, 2020–2026
<2min
Publish-to-searchable latency
Five production services ingesting, embedding, clustering and scoring your wire in real time, right now.

Capability tracks

Six ways to monetize the corpus you already publish

01 · Dedup

Story clustering

Collapses N wire hits covering one event — syndicated reprints, law-firm class-action pile-ons, multi-source earnings recaps — into one canonical story plus a coverage count. Cleans the feed and gives you a native "trending" signal for free.

Live: every article 2020–2026 carries a story id, size and canonical flag today.
02 · Taxonomy

Event classification

A 9-class taxonomy (earnings, M&A, legal, personnel, regulatory, capital action, product/partnership, promotional, other) trained on labeled samples — replaces or audits broad channel tags, and flags promotional/spam wire filler for automatic exclusion.

~80% held-out accuracy · 100% coverage 2020–2026
03 · Signal

Materiality & price attribution

Every article joined to split-correct forward returns (1d/3d/5d) and scored for P(the stock actually moved ≥5%). This is the piece a chatbot structurally cannot produce — it requires the full corpus and the price tape at once. Natural premium-tier differentiator.

0.68 AUC · flags top 18% of articles at 59% precision on real moves
04 · Search

Semantic search, full archive

Natural-language search over every sentence and article back to 2016 — "management commentary on margin pressure," not just ticker + keyword. Sits underneath any agent or product surface you want to expose it through.

Already the primary retrieval layer for one live downstream agent
05 · Aggregation

Corpus-wide trend surfaces

Questions no single article answers: which names had the most material news this quarter, event-type frequency by sector over time, stories that were under-covered but moved the stock anyway, and where realized returns diverged from predicted materiality (the market got surprised).

Computed once over the enriched set, refreshed continuously
06 · Graph

Entity & deal-flow linking

The same extraction technique we run in production elsewhere — pulling structured deal events (acquirer, target, amount, type) out of raw wire text at scale, cross-lingual — applied to your corpus turns "articles that mention a company" into a linked graph: company ↔ people ↔ deals ↔ every classified event, with a real timeline instead of a pile of headlines.

Proven at scale on comparable wire corpora: 6,300+ articles → 4,000+ structured deal records in one pilot

Why now

The moat that survives an LLM with web search

CapabilityA chatbot with web searchMeasured over the full corpus
"What's the latest on TICKER"Does this trivially
"Which of today's stories actually moved a stock"Cannot computeMateriality + return join
"What kind of event is this, systematically"Inconsistent, per-callFixed 9-class taxonomy
"Show me every M&A event for this company since 2020"Recency-biased, incompleteStructured, longitudinal timeline
"Which under-covered story just moved the stock 8%"Can't scan the archiveCoverage × materiality join

From article pile to linked graph

Track 06 is the connective layer over everything above: instead of a company's news being a flat list of headlines, it becomes a node connected to the people, deals and filings mentioned across the whole corpus — queryable as a timeline, not re-derived per chat session.

Where we'd start

  1. Live demo against your own archive on the call — this runs on your articles already, nothing to stand up.
  2. Pick one track to scope first: materiality/price-signal and story dedup are both fully built and the fastest to put in front of your subscribers.
  3. Entity/deal-flow linking (track 06) is the newest track — right fit for a scoped pilot on one vertical (e.g. M&A or funding announcements) before a full-corpus run.