News Narrative Tracker - Coverage To Narrative Read
Pricing
from $1,500.00 / 1,000 reports
News Narrative Tracker - Coverage To Narrative Read
Turn news coverage into a PR narrative read: main frames, repeated claims, disputes, spokespeople, and direction. Pay per report.
Turn news coverage into a PR narrative read: main frames, repeated claims, disputes, spokespeople, and direction. Pay per report.
What this Actor does
It collects public source data (see Evidence sources), then uses a language model to turn that raw material into one structured report. One run produces one report, and you pay only when a report is delivered.
Input
| Field | Type | Required | Description |
|---|---|---|---|
query | string | yes | Topic or brand |
maxRecords | integer | no | Articles to analyze |
time_period | string | no | Coverage window |
outputLanguage | string | no | Report language |
sampleData | boolean | no | Sample data (free, no data fetched) |
Advanced
| Field | Type | Required | Description |
|---|---|---|---|
llmApiKey | string | no | Optional. Leave empty to use the service's built-in analysis model. Supply your own OpenAI-compatible key to bill model usage to your account. Stored encrypted. |
llmApiBase | string | no | OpenAI-compatible base URL. Empty means the service default. |
llmModel | string | no | Model identifier. Empty means the service default. |
Output
One dataset item per run. Report fields:
| Field | Type | Description |
|---|---|---|
dominantFramings | array | Dominant framings |
repeatedClaims | array | Repeated claims |
contestedFacts | array | Contested facts |
voicesQuoted | array | Voices quoted |
toneTrend | array | Tone trend |
emergingAngles | array | Emerging angles |
responsePlan | array | Response plan |
Every report also carries:
| Field | Type | Description |
|---|---|---|
scope | object | Inputs and filters this report was produced for. |
sourceRecordCount | integer | Records collected from the source. |
recordsAnalyzed | integer | Records actually included in the analysis. |
recordsDroppedPastBudget | integer | Records excluded once the analysis budget was reached. |
chunksAnalyzed | integer | Number of analysis passes merged into the report. |
sources | array | Upstream dataset identifiers used as evidence. |
limitations | array | Known limits of this analysis. |
demoMode | boolean | true when no model key was configured. |
generatedAt | string | ISO-8601 timestamp. |
Evidence source
Upstream Actor: easyapi~google-news-scraper (Google News coverage). Data is collected at run time from publicly visible pages.
Pricing
Pay per event: $1.50 per delivered report (narrative-report). For comparison, Brand24 and media-monitoring platforms charge roughly $199-$699/mo.
Example input
Demo mode
If no model key is configured, the Actor still runs end to end and returns an extractive evidence report (record counts, frequency-ranked terms, verbatim samples) instead of model synthesis, with demoMode: true. This lets you evaluate a run before connecting a model.
Limitations
- Analysis is derived solely from publicly visible data collected at run time; it reflects the sampled window, not the full population.
- Results reflect the material collected in this run, not the entire population of available data.
- Analysis quality depends on how much public text the source exposes.
Independent third-party analysis service. Not affiliated with, endorsed by, or sponsored by the platforms whose public data is analyzed.
