Startup Funding Rounds Scraper - Global News

Fresh startup funding rounds from news in many countries and languages - company, amount in USD, round, investors, sector, HQ country - merged across every outlet that reported them, with an SEC Form D check for US rounds. For deal flow, B2B sales triggers and investor tracking. $8 per 1,000 rounds, AI extraction included.

What you get

• Round facts: company, headline (an English headline when one exists), round (pre-seed, seed, series-a ... series-h, growth, debt, grant, unspecified), amount + currency (as reported), amountUsd, valuationUsd, leadInvestors, investors, sector, hqCountry (ISO code), announcedAt (earliest article), confidence (0-1). null means unknown.
• Global, multilingual coverage: every Google News edition you pick (US:en, DE:de, FR:fr, IN:en...) is searched with funding wording in its own language - English, German, French, Spanish, Portuguese, Italian, Dutch, Polish and Ukrainian have their own phrases ("Finanzierungsrunde", "lève millions", "ronda de financiación"...); other languages use the English phrases. The AI returns English fields and the amount in USD, so a Spanish and a German report of the same deal end up in one clean row. In our test run (2026-11-04) 24% of the rounds appeared only in non-English local press.
• One round, many outlets: the same deal reported by Business Wire, TechCrunch and a local paper is one row, with every article in sources (outlet, domain, url, googleUrl, publishedAt, edition, headline, amount, currency), sourceCount, editions, firstSeenOutlet, localPressOnly, and amountConflicts when outlets report different amounts. In our test run 43% of the rounds were reported by 2+ outlets and merged.
• SEC Form D check: for US rounds (and rounds whose country is unknown) the Actor searches SEC EDGAR for a Form D filed by the company itself and adds confirmedByFormD, formDUrl and formDFiledAt.
• Free investors table (dataset investors): investor, rounds, asLead, companies, roundTypes, knownAmountUsd (sum of the known USD amounts of its rounds), most active first.
• New since your last run: roundId stays the same across runs; isNew and firstSeenAt mark rounds you have already received.

Use it for deal flow and market maps (every new round in your sectors and countries, every morning), B2B sales triggers (companies that just raised money and are about to hire and buy tools), investor tracking (which firms led the most rounds this week) and local press coverage (rounds in Germany, France, Spain, India or Poland that the English tech press never covered).

Sample output

Serval, reported on 2026-10-29 (trimmed, URLs shortened):

{
    "roundId": "r-197235dd8e44",
    "company": "Serval",
    "headline": "Serval Raises $150M Series C at $3B Valuation Led by Sequoia",
    "round": "series-c",
    "amount": 150000000,
    "currency": "USD",
    "amountUsd": 150000000,
    "valuationUsd": 3000000000,
    "amountConflicts": [],
    "leadInvestors": ["Sequoia"],
    "investors": ["Sequoia"],
    "hqCountry": "US",
    "announcedAt": "2026-10-29T12:49:26.000Z",
    "firstSeenOutlet": "Unite.AI",
    "sources": [
        {
            "outlet": "Unite.AI",
            "domain": "unite.ai",
            "url": "https://www.unite.ai/...",
            "publishedAt": "2026-10-29T12:49:26.000Z",
            "edition": "ES:es",
            "headline": "Serval recauda 150 millones de dólares en una ronda Serie C con una valoración de 3.000 millones para contratar a más de 500 personas",
            "amount": 150000000,
            "currency": "USD"
        },
        {
            "outlet": "Business Wire",
            "domain": "businesswire.com",
            "url": "https://www.businesswire.com/news/home/...",
            "publishedAt": "2026-10-29T13:00:00.000Z",
            "edition": "IN:en",
            "headline": "Serval Raises $150M Series C at $3B Valuation Led by Sequoia",
            "amount": 150000000,
            "currency": "USD"
        }
    ],
    "sourceCount": 2,
    "editions": ["ES:es", "IN:en"],
    "localPressOnly": false,
    "confirmedByFormD": true,
    "formDUrl": "https://www.sec.gov/Archives/edgar/data/2103406/000210340626000004/0002103406-26-000004-index.htm",
    "formDFiledAt": "2026-11-03",
    "isNew": true
}

Numbers from our test run

7 days, 6 editions (US:en, GB:en, IN:en, DE:de, FR:fr, ES:es), 2026-11-04:

What	Result
Headlines read	1,076 (500 unique)
Funding rounds found	188
Reported by 2+ outlets, merged into one row	81 (43%)
Only in non-English local press	46 (24%)
Lead investors named	75 of 188 (from headlines only)
Rounds with conflicting amounts between outlets	10
Form D: rounds checked / confirmed with a filing link	105 / 5

In a hand check of 50 headlines, company, amount, round and lead investor were all right in about 48 of 50 (96%).

How much does it cost?

$8 per 1,000 rounds. Example: the default input (US:en, last 2 days, 50 rounds) = 50 rounds x $0.008 = $0.40 per run (our test run, 2026-11-05).

• $0.008 per merged round: one round = one charge, no matter how many outlets reported it. AI extraction, USD conversion, publisher URLs and the Form D check are included; there are no add-on prices.
• The Investors table and RUN_SUMMARY are free.
• No start fee; platform compute is included.
• Max rounds (default 50) and your spending limit cap the cost. The Actor never decodes URLs, checks Form D or charges for rounds beyond your limits.

How to use

1. Pick the Google News editions (COUNTRY:language, up to 10) and the Time range.
2. Optionally filter by Round types, Minimum amount (USD), Sectors or HQ countries, and set Max rounds.
3. Click Start, then download JSON, CSV or Excel from the Output tab.

Main filters: editions, time range, round types, minimum amount, sectors, HQ countries.

Example input: {"editions": ["US:en", "GB:en", "DE:de", "FR:fr"], "timeRange": "7d", "rounds": ["seed", "series-a"], "minAmountUsd": 1000000, "maxRounds": 200}.

Use with AI assistants (MCP)

Connect this Actor to Claude, ChatGPT or any MCP client: https://mcp.apify.com?tools=steadywren/startup-funding-rounds-scraper

Try:

• "Find seed and Series A rounds of at least $2M announced in Germany, France and Spain in the last 7 days and list the lead investors."
• "Which investors led the most fintech rounds this week, and in which companies?"

How it compares

Checked against 7 funding-round Actors in Apify Store (READMEs and Store descriptions) on 2026-11-03.

	This Actor	Other funding-round scrapers in Apify Store
Non-English local press, translated and normalised to USD	yes (9 languages with own phrases)	not claimed (one partly: EU-Startups, Inc42)
One round merged from many outlets, conflicting amounts flagged	yes	not claimed
SEC Form D matched to the round (filing link)	yes	not claimed (one lists Form D filings as separate rows)
Investor activity table per run	yes, free	not claimed
Company enrichment (employees, LinkedIn)	no (sector + HQ country only)	several
Price per 1,000 rounds	$8	$2 - $80

Monitor daily

Run it daily with an Apify Schedule and "onlyNewRounds": true: each run delivers and charges only rounds you have not received yet. Send the results to Google Sheets, Slack, Make, Zapier, n8n or a webhook with Apify integrations, or read them through the Apify API (?format=csv or xlsx on the dataset URL).

How the history works: with Track rounds between runs on (default) the Actor saves the delivered rounds in a key-value store named funding-rounds-state in your account, under a key computed from your editions and filters (order and case do not matter; the key is in RUN_SUMMARY.stateKey). A round keeps its roundId across runs (company, round type and amount size), so a later run marks it isNew: false with its firstSeenAt. Rounds are forgotten 90 days after they were first delivered. Only new rounds delivers only rounds not delivered before and remembers them even with tracking off; rounds cut by Max rounds are not remembered and can come in a later run. With both off, isNew and firstSeenAt are null. Change the editions or filters and you start a fresh history.

FAQ

Is it legal? The Actor collects publicly available data only, without login and without personal data of private individuals. Check the site's terms for your use case. It reads public Google News feeds, public SEC EDGAR filings and the public ECB reference rates; names of people appear only as the headlines and investor lists published them.

What if Google News changes or a run fails? The run fails with a clear error instead of returning empty results, and you are not charged for results it did not deliver. No matching rounds finishes green with a message and nothing charged - but only when every feed was read and every headline judged. The run fails when no feed could be fetched (for example Google blocked the first requests), when the AI could not read any headline, or when no round was delivered and the result is incomplete (Google blocked or broke some feeds, or the AI stopped early). If Google starts refusing mid-run, the Actor stops making Google requests and delivers what it already has. Each run's summary (feeds, headlines, rounds found / delivered, Form D checks, AI cost, blocked flag, messages) is in the Storage tab -> Key-value store -> RUN_SUMMARY.

Do I need proxies? Are there limits? No proxies are used or needed. Google News returns at most 100 items per search, so when a search is full the Actor splits the time range into smaller date windows (down to single days, up to 300 headlines per search phrase and edition); a single day with more than 100 items still stops at 100. Requests are polite: Google requests start at least 0.5 s apart, with retries on 403/429/503 (1, 2, 4 s or Google's Retry-After up to 30 s). It never solves captchas, uses logins or rotates its identity. Only the headlines are read, so rounds announced only inside article text are missed. Up to 10 editions and 1,000 rounds per run.

How fresh is the data? Live at run time, from Google News of the chosen time range (today included). With the default input a run took about 1.5 minutes in our test run on 2026-11-04 (publisher URLs cached from an earlier run); a first run decodes every URL and took about 2.5 minutes on Apify (151 s, 50 rounds).

How are amounts normalised? amount and currency are the figure the outlets reported (ISO currency code), as a plain number ($1.2B = 1200000000, EUR 7,1 millions = 7100000). amountUsd and valuationUsd are converted with the latest ECB daily reference rates; USD amounts pass through unchanged. Indian units are computed by the Actor itself (1 lakh = 100,000, 1 crore = 10,000,000, so "Rs 5 crore" = 50,000,000 INR). "Up to $10M" is taken as the cap, 10,000,000. When no amount is stated the fields are null - the Actor never guesses numbers. If the ECB rates cannot be loaded, only USD amounts get amountUsd. When outlets report different amounts (more than 5% apart), the round takes the amount most outlets reported (ties: the earliest) and lists the others in amountConflicts with their outlets.

What does confidence mean? For each headline the AI gives a 0-1 confidence that it reports a real round with the right amount. The round's confidence is the average over its articles, +0.1 when 2 or more outlets reported it and -0.2 when there are amount conflicts, kept between 0 and 1. Low values are worth a quick look at sources.

Why are investors often empty? The Actor reads headlines (and Google's short snippet) only - it does not open article pages. Investors are filled only when a headline names them ("led by Sequoia"); in our test run that was 75 of 188 rounds. The AI is instructed to list only investment firms and funds; an individual named as an investor in a public headline (for example a known angel) can still appear in leadInvestors / investors.

How does the Form D check work, and what are its limits? For rounds with HQ country US or unknown, the Actor searches SEC EDGAR full-text search for Form D filings from 120 days before to 30 days after the announcement. It accepts only a filing by the company itself under the same name (legal forms like Inc. or LLC ignored); filings of funds and SPVs ("a Series of ...", "Fund", "LP") are rejected, and the newest match wins. A confirmed match also sets hqCountry to US. Limits: only US filings exist on EDGAR, the name must match exactly, and many startups file later or never. So confirmedByFormD: false means "no filing found in that window", not "fake round"; null means not checked (non-US company, Check SEC Form D off, or EDGAR refused and the checks stopped). In our test run, 105 rounds were checked and 5 confirmed.

Which editions and languages work? Any Google News edition written as COUNTRY:language, up to 10 per run. English, German, French, Spanish, Portuguese, Italian, Dutch, Polish and Ukrainian editions use funding phrases in their own language; any other language is searched with the English phrases. Whatever the language, the output fields are in English.

What is not included? No Crunchbase or PitchBook data, no employee counts, no founder or contact details, and no contact data of anyone (any person named in a headline stays in the original headline text). Data comes only from public news headlines, SEC EDGAR and ECB rates.

More from Steadywren

• Google News Scraper & AI Company Signals - Google News articles per company with AI event type, sentiment and trigger events.
• Substack Scraper & Newsletter Revenue Estimate - Substack leaderboards with an estimated revenue range per newsletter.
• Goodreads Scraper & Series Read-Through - Goodreads book series with ratings and read-through.
• All Steadywren Actors - small tools, steady results. Custom builds: steadywren.com.

Input reference

Field	Type	Default	Description
editions	array	["US:en"] (form prefilled with US:en, GB:en, IN:en, DE:de, FR:fr, ES:es)	Google News editions. Country and language of each Google News edition to search, as COUNTRY:language (US:en, GB:en, IN:en, DE:de, FR:fr, ES:es, BR:pt, IT:it, NL:nl, PL:pl, UA:uk ...). Each edition is searched with funding phrases in its own language; other languages use the English phrases. Up to 10 editions. Empty = US:en.
timeRange	string	2d	Time range. How far back to look for funding news (whole days, today included): 1d (today), 2d, 3d, 7d, 14d, 30d.
rounds	array	every type	Round types. Only these round types. Empty = every type.
minAmountUsd	integer	0	Minimum amount (USD). Only rounds of at least this many US dollars (other currencies converted with the latest ECB daily reference rates). Above 0, rounds without a known USD amount (not reported, or currency not convertible) are left out.
sectors	array	every sector	Sectors. Only rounds whose sector contains one of these words (English, case-insensitive, e.g. fintech, health, robotics). The sector comes from the headline, so rounds without one are left out. Empty = every sector.
countries	array	every country	HQ countries. Only companies headquartered in these countries (2-letter codes like US, GB, DE). The country comes from the headline, so rounds without one are left out. Empty = every country.
checkFormD	boolean	true	Check SEC Form D. For US companies and companies of unknown country, search SEC EDGAR for a Form D filed by the company itself under the same name (funds and SPVs are ignored) and add confirmedByFormD, formDUrl and formDFiledAt.
resolveUrls	boolean	true	Publisher URLs. Replace the Google News links of the first 3 sources of each round with the publishers' own article URLs.
trackChanges	boolean	true	Track rounds between runs. Remember the delivered rounds (per editions + filters) for 90 days and mark each round isNew with firstSeenAt.
onlyNewRounds	boolean	false	Only new rounds. Deliver (and charge) only rounds not delivered by an earlier run with the same editions and filters (works even with tracking off). Good for daily schedules.
maxRounds	integer	50	Max rounds. Stop after this many rounds (newest first, 1-1,000). You pay per round, so this also caps the cost.
maxAiCostUsd	number	0.5	Max AI cost per run (USD). Cap on the AI cost of reading headlines in one run (paid by the Actor, not charged to you; 0.05-2). When it is reached, the rounds read so far are delivered.

Search keywords: startup funding rounds, funding rounds, crunchbase alternative, startup funding, series a funding