Find Businesses That Just Got an SBA-Guaranteed Loan

For SBA loan brokers and bank business-development reps: get businesses that just had an SBA-guaranteed loan approved as one row each, with SBA's six separate FOIA files already merged.

Get one row for every business that just had an SBA-guaranteed 7(a) or 504 loan approved: who they are, where they are, who lent to them, how much, and what industry they're in.

Introduction

Every 7(a) working-capital loan and every 504 fixed-asset loan the U.S. Small Business Administration guarantees is disclosed at the individual-loan level in SBA's own public FOIA extract — because a lender only closes one of these loans when a business is actively investing or expanding, that disclosure is also one of the clearest "who is spending money right now" signals a small business can produce. This Actor reads SBA's own FOIA files live on every run and returns the loans approved in your chosen window: no scraping of a third-party mirror, no stale export.

Tutorial

1. Set Program to 7(a), 504, or Both.
2. Set Days back to how far behind SBA's own newest published record to look. SBA updates this file quarterly, so "days back" counts from the newest date SBA has actually published, not from today — see Limitations below.
3. Set Maximum rows to your spend cap. Every row bills, so this is also the most a single run can cost.
4. Optionally set Borrower state (e.g. TX) to one state, NAICS code prefix (e.g. 72 for accommodation and food service) to one industry, and/or Minimum loan amount to only see loans at or above a size worth your time.
5. Click Start. Results land in the dataset.

Pricing

$0.004 per row. No start fee.

A row here is a named business, its lender, and a specific dollar amount and date — not just "a filing exists." A run that returns 200 qualifying loans costs $0.80. A run that finds none because your filters are narrow costs nothing.

Maximum rows is the spend cap: the run cannot cost more than that number multiplied by $0.004, whatever the window holds.

Input and output examples

Input:

{"program": "both", "daysBack": 45, "maxItems": 25, "state": "TX"}

Output (one row, from a live run on 2026-09-15):

{
  "program": "7(a)",
  "sba_location_id": "0455644",
  "borrower_name": "Gallomar, LLC",
  "borrower_street": "297 W. ROUND GROVE RD",
  "borrower_city": "LEWISVILLE",
  "borrower_state": "TX",
  "borrower_zip": "75067",
  "lender_name": "Live Oak Banking Company",
  "lender_city": "WILMINGTON",
  "lender_state": "NC",
  "third_party_lender_name": null,
  "loan_amount_usd": 175000.0,
  "sba_guaranteed_amount_usd": 131250.0,
  "third_party_dollars_usd": null,
  "approval_date": "2026-06-30",
  "approval_fiscal_year": "2026",
  "first_disbursement_date": null,
  "term_months": 120.0,
  "naics_code": "811111",
  "naics_description": "General Automotive Repair",
  "franchise_name": "Midas",
  "project_county": "DENTON",
  "project_state": "TX",
  "business_type": "CORPORATION",
  "business_age": "Existing or more than 2 years old",
  "loan_status": "COMMIT",
  "jobs_supported": 7.0,
  "source_url": "https://data.sba.gov/sites/default/files/uploaded_resources/FOIA_7a_FY2020_Present_asof_260630.csv",
  "fetched_at": "2026-09-15T17:02:24Z"
}

Every field in a row:

Recommendations

Run weekly with Days back set to 90 to reliably cover SBA's quarterly publishing cycle without missing a batch, and use Minimum loan amount to focus on loans large enough to be worth a call — a $5,000 SBA Express line and a $3.2 million 504 debenture are very different prospects.

Limitations

Not same-day. SBA republishes this file on its own quarterly schedule. The newest row available as of 2026-09-15 was dated 2026-06-30 — about an 11-week lag. Days back is therefore counted from the newest date SBA has actually published, not from today, so a small window still returns rows instead of silently coming back empty between SBA's updates.

Recently committed, not always funded. Many rows carry

loan_status: "COMMIT"

One record source, not per-loan pages. SBA publishes this as a bulk file rather than one web page per loan, so source_url names the exact file this row came from rather than a unique page for that specific loan. sba_location_id is SBA's own identifier for the record if you need to reference it precisely.

Only what SBA states. No amount is estimated and no field is inferred; a blank field in SBA's file is returned as null.

FAQ

Where does the data come from? SBA's own open-data site, data.sba.gov, which publishes the same loan-level FOIA extract it releases in response to Freedom of Information Act requests.

Does it need an API key or a login? No. Nothing here requires an account, key, or registration of any kind.

Why did I get fewer rows than Maximum rows? Either your window and filters matched fewer loans than your cap, or (rarely) this Actor's per-run data-fetch limit was reached before your filters were fully satisfied — check OUTPUT.warnings.

What happens if SBA's site is down? The run pushes no rows, costs nothing, and writes a status of UNKNOWN with the error to OUTPUT. It never fills the gap with made-up rows.

Is there support? Issues can be raised through the Apify Store issues tab for this actor. No response time is promised.

Refunds and support

Refunds: if a run returns zero rows for a valid input, email operations@ustechautomations.com within 7 days and we refund that run. Support: same address, replies within 2 business days.