NSF Awards, Grants & Research Funding Bulk Scraper

Export National Science Foundation awards from official fiscal-year bulk archives. Unlike the interactive NSF award search, which limits immediate search exports to 3,000 results, this Actor streams complete annual archives and can combine multiple years into one normalized dataset.

Each award includes funding amounts, dates, recipient institution and performance location, principal investigators, program elements, directorate and division, program officer, public abstract, and official award link. Current annual archives contain thousands of records; multiple years provide tens or hundreds of thousands.

What you can use it for

• Discover recently funded researchers, laboratories, universities, nonprofits, and technology companies.
• Build lead lists for scientific equipment, software, cloud, recruiting, publishing, consulting, and research services.
• Track research topics, emerging technologies, funded institutions, and geographic funding distribution.
• Find awards by keyword, directorate, state, recipient, amount, activity, or fiscal year.
• Analyze principal-investigator and institutional networks.
• Power grant intelligence, academic research, policy analysis, and market mapping.

Data included

• NSF award ID, fiscal year, title, public abstract, agency, transaction type, award instrument, and CFDA number
• start and expiration dates, current award amount, total intended award amount, and ARRA amount
• awardee institution and legal name, address, telephone, UEI, parent UEI, and congressional district
• project performance location
• principal investigator names, roles, public award contact emails, and NSF identifiers
• NSF directorate, division, program elements, and program references
• program officer name, email, and telephone
• obligations by fiscal year where supplied
• direct official award and bulk-source links

Blank numeric fields remain null. Identifiers, postal codes, and congressional districts remain strings. Disable includeAbstract when you need a smaller high-volume dataset.

Input

The default run automatically scans the current and previous fiscal-year archives and returns up to 1,000 awards. maxResults supports up to 500,000 across the selected years.

• years: one or more available NSF fiscal years
• states: recipient institution state codes
• awardeeContains: recipient institution or legal-name fragment
• keyword: title, abstract, or program-element keyword
• directorates: current code or name fragments such as CSE, ENG, BIO, or MPS
• minimumAwardAmount: minimum total intended award amount
• activeOnly: expiration date has not passed
• includeAbstract: include or omit full abstracts

Example: active AI-related awards in Computer and Information Science and Engineering (current bulk code CSE) and Engineering:

{
  "maxResults": 10000,
  "years": [2026, 2025, 2024],
  "keyword": "artificial intelligence",
  "directorates": ["CSE", "ENG"],
  "activeOnly": true,
  "minimumAwardAmount": 100000
}

Example: California university awards without large abstracts:

{
  "maxResults": 25000,
  "years": [2026, 2025, 2024],
  "states": ["CA"],
  "awardeeContains": "university",
  "includeAbstract": false
}

Output example

{
  "awardId": "2625972",
  "fiscalYear": 2026,
  "title": "Conference: Label-Free Single Molecule Sensing",
  "totalIntendedAwardAmount": 10000,
  "startDate": "2027-03-15",
  "expirationDate": "2027-08-31",
  "awardee": { "name": "Gordon Research Conferences", "state": "RI", "uei": "XL5ANMKWN557" },
  "principalInvestigatorNames": ["Georgios Alexandrakis"],
  "directorateCode": "ENG",
  "divisionCode": "CBET"
}

The default Apify dataset can be downloaded as JSON, CSV, Excel, XML, or RSS and consumed through API clients, webhooks, Zapier, Make, Google Drive, and other integrations.

Performance and reliability

On each run, the Actor requests the official NSF archive directory and receives fresh signed download links. It downloads selected fiscal years newest first, streams each ZIP entry, parses one JSON award at a time, applies filters, and writes batches. Annual ZIPs are never unpacked wholesale to local storage or held as one large object.

This design bypasses the interactive search's 3,000-record display/export ceiling without working around access controls: it uses the NSF's documented complete fiscal-year downloads. Missing years are logged and skipped; the run fails clearly if nothing matches.

Data source, public domain, and responsible use

The source is the official NSF Award Search annual bulk data. NSF states that posted award data, including award abstract text, is public domain. Field availability varies by era, and older fiscal years can omit abstracts or modern identifiers.

Funding amounts and dates can change through amendments. Use the linked official award page for current grant facts. Public investigator contact information should be used lawfully and professionally; follow applicable privacy, anti-spam, and marketing rules. This Actor is not affiliated with or endorsed by NSF.

Pricing

Pay-per-event pricing charges a small start event and one result event per saved award. maxResults is the charge ceiling for results. Downloaded or scanned awards that fail filters are not charged as result events.

Changelog

2026-08-03 — 1.0.0

• Initial release.
• Verified the new post-January-2025 JSON annual archive format and current 2026 archive contents.
• Added multi-year bulk processing beyond the 3,000-result search cap, streaming JSON/ZIP parsing, grant and recipient filters, optional abstracts, retries, batch writes, and dataset overview.