Extract data from 1099-NEC forms

Since the IRS moved nonemployee pay back onto its own form in 2020, the 1099-NEC has become the document every accounts-payable and tax-prep team handles in volume each January. A company that paid 400 contractors issues 400 of these, and the recipients hand them to a preparer alongside their other income documents. The form is compact but unforgiving: Box 1 reports nonemployee compensation, Box 4 any federal income tax withheld under backup withholding, and Boxes 5 through 7 the state tax, state identifier, and state income. The payer name and TIN at the top and the recipient TIN below tie the form to both parties for matching against the IRS information-return system. The difficulty is identity and reconciliation rather than layout. The payer TIN follows the two-digit-then-seven-digit EIN format, while a recipient TIN may be an EIN or a masked SSN, and the two must not be confused. Box 1 has to reconcile against what the payer's AP ledger recorded for that vendor across the year, and a mismatch is the most common 1099 correction. A corrected 1099-NEC carries a checkbox that changes how it supersedes the original. State boxes repeat when a contractor worked across two states, so a single recipient can have two state lines. Talonic reads the 1099-NEC and returns payer, recipient, both TINs, the Box 1 compensation, any backup withholding, and the state lines as discrete fields. The corrected-form checkbox is captured, and amounts come back as clean numbers so AP can reconcile them against the vendor ledger without re-keying the form.

What gets extracted from 1099-NEC forms

How extraction works for 1099-NEC forms

1099-NEC forms reach a preparer as issuer PDFs, payer-portal downloads, and recipient phone photos, and the official IRS layout is consistent enough to anchor on box numbers. Talonic reads the form against the 1099-NEC box map in the Field Registry, which binds each value to its numbered box. Payer TIN and recipient TIN are captured as separate fields and validated against the EIN format so they are never swapped. Box 1 nonemployee compensation returns as a clean number for reconciliation against the accounts-payable ledger, and the corrected-return checkbox is captured because it determines whether the form supersedes a prior filing. Repeated state boxes are kept as separate state lines. Every value returns with a confidence score and pixel-region provenance under DIN SPEC 91491 conformity, so a US preparer or AP analyst can verify a box against the source before filing or matching.

Sample extraction

A 2025 IRS Form 1099-NEC issued to a single contractor

{
  "tax_year": 2025,
  "payer_name": "Northgate Logistics Inc.",
  "payer_tin": "12-3456789",
  "recipient_name": "Daniel Okafor",
  "recipient_tin_last4": "1042",
  "box1_nonemployee_compensation": 42800,
  "box4_federal_withholding": 0,
  "state_lines": [
    {
      "state": "CA",
      "payer_state_no": "123-4567",
      "box5_state_tax": 1284,
      "box7_state_income": 42800
    }
  ],
  "corrected": false
}

Frequently asked