See how coverage of a public figure, place, company, or event develops day by day, and where each channel runs warmer or cooler than a balanced reference built from publishers across outlet types and political perspectives.
Current channels: Google News editorial algorithm, financial press (Finnhub), and local outlets across all 50 US states. More channels — social media, newsletters — coming soon.
// Narrative time series for Federal Reserve
GET /v1/narratives
?entity=federal+reserve
&concept=rate+hike
&channel=financial
&from=2026-09-01
{
"entity": "federal reserve",
"concept": "rate hike",
"channel": "financial",
"series": [
{
"date": "2026-09-14",
"mentions": 47,
"sentiment": -0.220,
"bdms": -0.065,
"bdmsCiLow": -0.180,
"bdmsCiHigh": 0.050,
"bdmsSampleSize": 18,
"bdmsComputedAt": "2026-09-12",
"cnd": -0.155
}
// ... 89 more days
]
}Narrative endpoints compare channel sentiment with Balanced Discoverable Media Sentiment (BDMS): a publisher-balanced reference with a 95% confidence interval. The subject of coverage — a person, company, place, or event, which we call an entity — is tracked alongside the specific storyline around them, which we call a concept. Together they give you daily narrative history: 7 days on Free, 90 on Starter, and everything since March 2026 on Professional.
/v1/narrativesDaily time series for any entity+concept pair: mention count, channel sentiment (Google News, financial, local, or all), the BDMS reference with its 95% CI, and the divergence between them (CND). 7 days of history on Free, 90 on Starter, everything on Professional.
/v1/trendingEntities ranked by recency-weighted media velocity. 3× weight for last 24h, 2× for 24–48h. Filter by category.
/v1/entity/{entity}/coverageAll concepts associated with an entity — every angle the media is framing them through. Mention count, observed sentiment, and the BDMS reference per concept.
/v1/searchResolve canonical entity names. "TSA" → "transportation security administration". Every other endpoint requires the canonical form.
/v1/articlesThe articles behind a storyline, an entity or a keyword, newest first: URL, title, source, date, channel, state for local outlets, and per-article sentiment with the BDMS reference as of each date. History follows your plan; Free keys see up to 5 articles from the past week.
/v1/stocks/sentimentFinancial press sentiment for the most-covered stocks, with market-wide context. Starter gets the top 5; Professional gets up to 20 with daily sentiment and price trends.
/v1/local/{state}State-level trending coverage for all 50 US states. Top articles with entity-concept sentiment, 24-hour window.
/v1/analyzeSend any article text. Returns every entity mentioned with its sentiment, plus the BDMS reference and CI bounds. You scrape, we signal.
/v1/enrichCompute the BDMS reference on demand for any entity+concept pair not yet in the dataset, with its CI and contributing publishers.
Two signals per entity+concept pair: what each channel carries, and a balanced reference built from discoverable media.
We ingest from multiple channels: Google News section feeds (editorial algorithm), financial press via Finnhub, and local RSS outlets across all 50 US states. Each channel has its own sentiment for any entity+concept pair, so you can compare how different channels cover the same story.
For each pair, we search Google News for the past week of coverage and sample up to four articles per outlet type (wire, broadcast, print, independent, and local), rotating across left, center and right publishers.
An LLM scores each article for its sentiment toward the entity regarding that concept. Not topic-level tone, but sentiment about one entity through one lens.
We average within each publisher, then across publishers within each outlet type, then equally across outlet types. High-volume outlets count once, and every outlet type gets an equal say. The result is Balanced Discoverable Media Sentiment (BDMS).
Resampling publishers 1,000 times within each outlet type gives a 95% interval. A narrow interval means publishers cover the pair in a similar tone; a wide one means coverage is genuinely mixed.
CND is channel sentiment minus BDMS, tracked day by day. See where each channel runs warmer or cooler than the balanced reference, and how that gap moves with the news cycle.
Every plan includes the narrative time series, all three channels, BDMS and divergence, plus search, trending and entity coverage. Upgrade for more history and calls, stock sentiment, local news and article analysis.
Free tier, no credit card required. 1,000 calls per month to explore the data.
Bug report, feature request, or just a thought — we read everything.