Homeowner experiences, agent discussions, E&S/surplus lines, and FAIR Plan coverage in California wildfire zones
Audience labels and classification rules have moved to the Audiences tab.
Set maximum results per scrape run for each active source. These limits are advisory — scripts will read them but won't hard-block.
| Source | API | Cost | Max Results / Run |
|---|---|---|---|
| reddit.com | Reddit Scraper Lite | — | |
| serpapi.com | SerpAPI (search and discovery) | $0.015/search | |
| twitter.com | Tweet Scraper V2 | $2.00/1K |
Customize the AI prompt used when generating analysis reports from extracted content. Use {{articles}}, {{project_name}}, and {{report_type}} as template variables. Leave blank to use the default prompt.
Configure how the word cloud extracts, scores, and displays terms from your posts. Each section has a brief explanation and a “Learn more” link with details.
Words added here are excluded from the word cloud, in addition to ~200 built-in English stopwords.
Multiply a term's weight to push it higher in the word cloud ranking.
Controls how multi-word phrases (bigrams and trigrams) are identified and filtered.
Groups inflected word forms together (e.g., “policy” and “policies” become one entry).
Controls how many terms appear and the minimum frequency required.
Controls the sensitivity of the up/down trend arrows shown on terms in the cloud.
These features are controlled via the word cloud UI, not this settings page. Expand for details.
The word cloud has several interactive features available directly on the analytics page.
Archive this project
Hide from main project list. Can be restored later.
Delete this project
Permanently delete this project and all its data. This cannot be undone.