Crunchbase News & Funding Coverage Scraper

Crunchbase's editorial newsroom publishes thousands of articles about funding rounds, acquisitions, IPOs, layoffs and startup trends. This actor turns that public archive into structured dataset rows: full article text, authors, categories, tags, dates, media metadata, SEO fields and funding hints. It works without an API key, login, cookie or proxy.

What you can collect

• Latest news — the newest articles, optionally narrowed by keyword, category, tag, author or date.
• Keyword search — full-text search across headlines and article bodies; choose Search relevance in Sort by to rank matches first.
• Funding coverage — the Funding reports section, with optional keyword and date filters.
• By category — one or many category slugs, names or IDs (for example venture, ai, ma, ipo).
• By tag — one or many tags (for example unicorn, seed-funding, venture-funding).
• By author — one or many Crunchbase News bylines.
• By article URLs or IDs — paste article links, numeric IDs or slugs and fetch exactly those articles.

Input

Field	Type	Default	What it does
mode	select	latest	Pick one of the seven collection modes above
search	string	–	Keyword matched against titles and body text
categories	array	–	Category slugs, names or IDs; OR-combined
tags	array	–	Tag slugs, names or IDs; OR-combined
authors	array	–	Author slugs, names or IDs; OR-combined
urls	array	–	Article URLs, numeric IDs or slugs (By article URLs or IDs mode)
publishedAfter	string	–	Lower date bound (YYYY-MM-DD or ISO 8601)
publishedBefore	string	–	Upper date bound (YYYY-MM-DD or ISO 8601); a date-only value includes that whole day
orderBy	select	date	date, modified, title or relevance
order	select	desc	desc or asc
includeContent	boolean	true	Include the full article body in content
maxItems	integer	50	Maximum rows (1–10,000)

All filters except urls can be combined in the list modes. The By article URLs or IDs mode returns exactly the requested articles and ignores the list filters.

Examples

Latest AI funding articles from 2026:

{
  "mode": "funding",
  "search": "AI",
  "publishedAfter": "2026-01-01",
  "maxItems": 100
}

Keyword search with full text:

{
  "mode": "search",
  "search": "Series A",
  "includeContent": true,
  "maxItems": 50
}

Browse two categories:

{
  "mode": "byCategory",
  "categories": ["venture", "artificial-intelligence"],
  "orderBy": "date",
  "order": "desc",
  "maxItems": 200
}

Fetch specific articles:

{
  "mode": "byUrls",
  "urls": [
    "https://news.crunchbase.com/venture/biggest-startup-rounds-ai-defense-space-tech-shield-zipline/",
    "93501",
    "vertical-ai-based-customer-service-flip-raise"
  ]
}

Old saved tasks keep working: legacy keys such as query, category, after and maxResults are mapped onto the current fields inside the actor, not shown in the form. search and maxItems are the current names.

Output

Every row is omit-empty: a value the source does not publish is left out instead of being written as null, "" or [].

Article identity and text

id, url, slug, title, excerpt, content (when includeContent is on), publishedDate, publishedDateGmt, modifiedDate, modifiedDateGmt, scrapedAt.

Authors

authorId, authorName, authorSlug, authorUrl, authorIds, authorNames, authorSlugs.

Taxonomy

categoryIds, categoryNames, categorySlugs, tagIds, tagNames, tagSlugs, isFeatured.

Media

featuredImageId, featuredImageUrl, featuredImageAlt, featuredImageCaption, featuredImageWidth, featuredImageHeight, featuredImageMimeType, featuredImageThumbnailUrl.

SEO and reading metadata

metaDescription, canonicalUrl, keywords, wordCount, readingTimeMinutes, inLanguage.

Funding hints

roundTypes, fundingAmounts, mentionedOrganizations, mentionedPeople.

roundTypes and fundingAmounts are deterministic text hints extracted from the headline and body. They help triage funding coverage; they are not audited Crunchbase funding-round records. mentionedOrganizations and mentionedPeople come from the Crunchbase profile links embedded in the article body.

How pagination works

The actor reads the public WordPress REST API 100 articles at a time and keeps walking pages until maxItems is reached or the archive is exhausted. The archive currently holds about 8,800 articles, so maxItems above that returns the whole archive. Results are de-duplicated by article ID.

Use cases

• Deal-flow monitoring — watch new funding coverage by keyword, category or author and forward the rows to a CRM or alerting job.
• Sector research — pull the venture, AI, fintech, climate or health archive for a market map.
• Competitive intelligence — search for a company, investor or product across years of coverage.
• PR and media monitoring — track new mentions of your company or portfolio.
• Content and newsletter workflows — use excerpts, categories and reading time to assemble a digest.
• Dataset building — export clean JSON, CSV or Excel for text mining, embeddings or dashboards.

FAQ

Do I need a Crunchbase account, API key or cookie? No. The actor reads Crunchbase News, the public editorial site, with no credentials and no proxy configuration.

Does this return Crunchbase company profiles or structured funding rounds? No. The main Crunchbase company database is login-gated, and its public organization pages hide funding amounts and dates from anonymous visitors. This actor covers the editorial news archive. The roundTypes, fundingAmounts, mentionedOrganizations and mentionedPeople fields are text and link hints from those articles, not the company database.

How far back does the archive go? About 8,800 published articles, with the earliest posts dating back to the site's early years. The exact count grows as new articles are published.

Why is content missing on some rows? Either includeContent is turned off, or the source returned an empty body for that article. The excerpt, title and metadata are still returned.

Why do some rows have no tags or no coauthor IDs? Older articles predate the current tag and coauthor setup. The byline, categories, image and SEO fields are still filled from the article itself.

Can I combine search, categories, tags, authors and dates? Yes, in every list mode. The filters are combined with AND across filter types and OR inside each array.

What is the difference between publishedDate and publishedDateGmt? publishedDate is the site's local time (America/Los_Angeles). publishedDateGmt is the same instant in UTC.

Is the actor affiliated with Crunchbase? No. It is an independent, third-party tool that reads publicly published Crunchbase News articles. It is not endorsed by or affiliated with Crunchbase.

Scope and limitations

• Editorial news only. Company profiles, investor profiles, structured funding rounds, valuations and contact data from the main Crunchbase database are not included.
• Anonymous view limits. Crunchbase hides funding amounts and dates on public company pages behind an obfuscated view; this actor does not attempt to bypass that.
• Archive ceiling. The WordPress API currently exposes about 8,800 posts; maxItems above that returns the whole archive.
• Text hints are heuristic. Round types and dollar amounts are regex extractions from prose. A single article can mention several amounts (round size, valuation, investor commitments), so use the raw values and the article link for verification.
• Partner content excluded. Sponsored partner_content posts are a separate, older post type and are not mixed into news results.
• No wall-clock cap. The actor stops on maxItems, source exhaustion or a source error, so a large run returns the requested rows rather than a time-limited subset.

Data source

Public WordPress REST API at https://news.crunchbase.com/wp-json/wp/v2. The actor sends a normal browser user agent, retries transient 429/5xx responses with backoff, and requires no user-supplied credentials.