I-9 Form Parser That Reads Every Field, Handwriting Included

Key Takeaways

• An I-9 form parser reads a completed Form I-9 and turns every field, checkboxes and hand-printed boxes included, into structured data. No template to build, nothing typed twice.
• Parseur covers all four parts of the form: the Section 1 attestation, the Section 2 List A, B and C documents and employer certification, the Supplement A preparer details, and Supplement B reverification.
• The same mailbox takes the rest of the paperwork that lands on HR and finance desks: W-9s, W-4s, W-2s, Form I-94, visas and verification of employment requests.
• Low-confidence fields get flagged for a person and corrected inside Parseur, so the Social Security number and the expiration date pass human eyes before anything moves.
• Parseur extracts. Zapier, Make, Power Automate, n8n, a webhook or the API delivers, and that step writes the record into your HRIS. There is a free plan, so you can run your own worst scans through it before you pay anyone anything.

Nobody was hired to retype an I-9. It happens anyway, because the form arrives as a scan, the data has to live in the HRIS, and the only path between them runs through a keyboard.

The stakes are not spread evenly across the onboarding packet. A typo in an emergency contact is an inconvenience. A wrong document expiration date on an I-9 is a missed reverification, and an unsigned form is a finding. According to SHRM, paperwork violations on Form I-9 can result in fines ranging from $234 to $2,332 per incident, while knowingly hiring unauthorized workers can cost between $583 and $4,667 for first-time offenses.

So the form that most deserves careful handling gets the tiredest method available, at the end of a stack, from someone who has read forty of them today. This is not a volume problem you outgrow either. Sixty hires a year is sixty chances to mistype a date that an auditor reads back to you three years later.

What follows is the whole form field by field, what Parseur does with the values it is unsure about, where the data goes after that, and what it costs to find out whether any of it survives your scans.

What is an I-9 form parser?

An I-9 form parser is software that automatically reads a completed Form I-9, Employment Eligibility Verification, and converts every field on it into structured data. It captures the employee attestation, the identity and work authorization documents recorded in Section 2, the employer certification and any reverification, and returns them as CSV, Excel or JSON that HR, payroll and compliance systems can use directly.

Some vendors sell the same thing as an I-9 form data extractor. Same job, same output.

Parsing is not scanning. A scan gives you a picture of a form, which is fine for the file and useless for everything else. Parsing gives you the values: a name you can search, an expiration date you can set an alert on, a citizenship category you can report on. One is storage. The other is data.

Three things make the I-9 hard for ordinary text recognition. It is almost always hand-printed. Its load-bearing values are checkboxes rather than words. And it points at a second set of documents, the List A, B or C identity papers, whose details get transcribed onto the form and then, in most companies, transcribed again into a system.

What Parseur pulls off an I-9, field by field

Here is the whole form, section by section, as Parseur returns it.

Section 1, employee information and attestation:

• Last name, first name, middle initial, and other last names used
• Address, apartment number, city, state and ZIP code
• Date of birth
• U.S. Social Security number
• Employee email address and telephone number
• Citizenship or immigration status, the attestation checkbox: citizen, noncitizen national, lawful permanent resident, or noncitizen authorized to work
• USCIS number, A-Number, Form I-94 admission number, or foreign passport number and country of issuance, whichever the attestation requires
• Work authorization expiration date, when the status carries one
• Employee signature date

Section 2, employer review and verification:

• List A document title, issuing authority, document number and expiration date, for a document establishing both identity and work authorization
• List B document title, issuing authority, document number and expiration date, for identity
• List C document title, issuing authority, document number and expiration date, for work authorization
• Additional document rows, for the second and third entries a List A document can carry
• Additional information box, including alternative procedure notations for remote document examination
• First day of employment
• Employer or authorized representative signature date, printed name and title
• Employer business or organization name and address

Supplement A, preparer and translator certification:

• Name and address of each preparer or translator
• Certification signature dates

Supplement B, reverification and rehire:

• Rehire date, when an employee returns within three years
• New name, when it changed
• Reverification document title, number and expiration date
• Employer signature date for each reverification event

That last block is the one nobody thinks about until it matters. Reverification is a calendar problem, and a calendar problem needs a date in a system, not a date in a filing cabinet.

Why ordinary OCR fails on an I-9

I-9 form OCR has to survive two things a printed-document engine was never built for.

The first is handwriting. Employees fill in Section 1 by hand, usually in comb boxes, sometimes in ballpoint on a form that was photocopied twice. Reading that reliably takes intelligent character recognition rather than plain OCR, and that is the work Parseur's Vision AI engine does on a scan or a phone photo.

The second is checkboxes. The citizenship attestation and the List A, B or C selection are not sentences, they are marks in boxes, and they carry more legal weight than anything else on the form. An I-9 with the wrong citizenship box ticked is an audit finding no matter how cleanly the name was read. Parseur reads the marks as values, not as smudges to be ignored.

W-9s, I-94s and the rest of the government paperwork

The I-9 rarely arrives alone, and the same mailbox handles what comes with it.

Form W-9, for contractors and vendors. The form that most often arrives alongside the I-9, and typed far more often than hand-filled, which makes form W-9 OCR the easy half of the job. Point the same mailbox at it and Parseur runs as a form W-9 parser too, returning:

• Legal name, and business or disregarded entity name
• Federal tax classification, including the LLC tax classification letter
• Exemption codes and FATCA reporting code
• Address, city, state and ZIP code
• Account numbers listed on the form
• Taxpayer identification number, SSN or EIN
• Certification signature date

That is the whole set finance needs before 1099 season stops being a scramble. If tax paperwork is the bigger half of your problem, the tax data extraction tools roundup covers the wider category.

Form I-94 and travel documents. Arrival and departure records, visas, passports and permanent resident cards, with the document numbers, issuing authorities and expiry dates that drive work authorization tracking. Immigration teams run this across whole case files, covered in AI immigration document extraction.

W-4 and W-2. Filing status, dependents, adjustments, withholding elections and wage figures, feeding payroll rather than compliance.

Verification of employment requests. The letters lenders and background check firms send asking you to confirm dates and title. A verification of employment form parser reads the request itself, so answering it is a review rather than a retype.

One boundary worth drawing. If your problem is the whole new hire stack, the W-4 and the direct deposit form and the benefits sheet landing together, that workflow lives on onboarding document data extraction. This page is for the forms where the stakes, not the volume, make the typing dangerous.

What retyping an I-9 actually costs

Manual handling is still the norm, and it fails in predictable places. Flair HR reported that nearly 40% of the onboarding journey consists solely of paperwork tasks, which slows every start date and multiplies the chance of a data error.

Formtek put the shift at up to 80% of companies adopting automated document processing by 2026. Judging by where the errors turn up, the I-9 has a good claim to going first.

Where it breaks:

• Typing into the HRIS. Staff read field by field and key field by field. A transposed digit in a Social Security number surfaces weeks later, in payroll or in an audit.
• Formats that never match. Scanned PDFs, email attachments, printed copies, e-signed files, and a photograph taken at an angle on a remote hire's kitchen table.
• Expiration dates nobody is watching. Work authorization expires. If the date only lives on paper, reverification is something you remember late.
• Gaps that sit there quietly. Nobody spots an unsigned form or an unticked attestation until someone goes looking, and by then it has been in the file for a year.
• No way to say what has arrived. Which forms came in, which were entered, which are still owed. Inbox folders and a shared spreadsheet stop scaling at a few hires a week.

The evidence on error rates is not subtle. A detailed overview by Lightico found that up to 76% of paper I-9 forms contain errors, and that electronic systems with smart validation and real-time checks substantially cut those mistakes. In practice, platforms like Imagility report 65% faster I-9 processing times and 100% compliance after automating the step.

How Parseur's I-9 form parser works, from inbox to HR system

Parseur customers saved up to 152 hours of manual data entry every month in 2025, roughly $7,000 in labor costs, or more than $80,000 a year. Five steps get you there, and there is no template to build and nothing to configure before the first form works.

1. Send in the forms you already receive

Every Parseur mailbox has its own email address. Forward signed I-9s, W-9s and supporting IDs to it, set your onboarding inbox to forward them automatically, or push them in through the API. Scans, digital PDFs and mobile uploads land in one place instead of five people's inboxes.

2. Extraction starts on arrival

Parseur's Vision AI engine reads PDFs, scans and photos, and its Text AI engine handles emails and text documents. Layout is not a constraint, so the 2025 edition of the form, the 2023 edition still sitting in your files, and a photographed copy of either all come out the same way.

3. Review the two fields that hurt

Every field comes back with a confidence signal, so low-confidence values can be flagged for a person and everything else can flow. Correct a flagged field in Parseur and the corrected value is what leaves Parseur, so a bad read never reaches the systems downstream. Most teams check the Social Security number and the document expiration date, approve, and move on. A minute per hire, and it is the difference between a fast process and a fast mistake.

4. Get data that needs no cleanup

Results come back as CSV, Excel or JSON, formatted identically every time and ready for reporting, compliance checks or system import. Extraction reaches accuracy of up to 99.9% on structured documents in Parseur's internal testing and customer reporting, with the validation step covering the rest.

5. Route it where it belongs

Parseur extracts the fields, then hands them to Zapier, Make, Power Automate, n8n, a webhook or its API, and that step writes the record into your HRIS, ATS, payroll system or secure storage. There is no native HR connector here and no pretending otherwise, because the automation platform is what talks to your systems. Map each field to its counterpart once, when you build the connection, and the flow runs unattended after that.

The security question arrives before the accuracy one

One I-9 packet holds a Social Security number, a date of birth, a government ID number and often a photograph of the ID itself. Very little else your company moves by email carries that much in one envelope, so the question is fair, and it deserves an answer with names in it rather than a reassurance.

Parseur is SOC 2 Type II compliant, and the report is available on request. Documents are encrypted in transit and at rest, and access runs on role-based permissions with audit trails, so you can say who saw which form and when. Parseur is GDPR-native and meets California CCPA requirements, and your documents are never used to train AI models and never sold.

Retention is the setting to look at first on this form. Set a window and documents delete themselves when it runs out, which matters more on I-9s than on anything else, because retention here is a legal requirement rather than a housekeeping preference.

Two things worth building into the workflow from the start:

• Keep the review step on the fields that carry the risk. The Social Security number and the work authorization expiration date should pass a person before they reach anything downstream.
• Store the original next to the extracted values. An I-9 audit asks to see the document, the data, and when each was recorded. Keep the three together and the audit is a lookup.

Extraction is not a substitute for the parts of I-9 compliance that are yours. You still examine the documents, complete Section 2 within three business days, and retain the form for three years after hire or one year after termination, whichever is later. What automation removes is the retyping, and the errors that ride along with it.

The desks these forms land on

HR teams at fast-growing companies

Structured onboarding programs improve new hire retention by 82% and productivity by over 70%, according to SHRM. No program gets there while week one is a typing exercise. Parseur extracts the fields the moment forms arrive, so payroll, benefits and employee records start from data instead of a queue of PDFs. Deel puts the ceiling on it: software tools can reduce onboarding time by up to 80%, with HR staff spending as much as 57% of their time on administrative work that automation can absorb.

Compliance and legal teams

The audit does not care how busy the week was. Parseur builds the digital trail as the forms arrive rather than reconstructing it a year later, with expiration dates sitting somewhere an alert can reach them.

Staffing agencies and PEOs

High volume, fast turnover, hundreds of forms a month across client accounts. One workflow covers many employers here, because Parseur reads each client's paperwork the same way regardless of layout. FlowForma reports that 87% of organizations using onboarding automation see faster time to productivity for new hires, and as Rinku Thakkar noted, 68% of recruiting professionals believe investing in new recruiting technology is the most effective way to boost performance, with a 16% increase in new hire retention when onboarding tasks are automated.

Remote-first organizations

Nobody touches the paper. Every form arrives as a scan or a phone photo at whatever quality the hire's kitchen lighting allowed, and Parseur reads both, so capture stays consistent across time zones. Compunnel noted that organizations with structured remote onboarding report 82% higher new-hire retention, and The Blueprint AI reports Remote.com cutting time spent on manual HR tasks by 97% after connecting its HR systems through automation.

Where the data goes next

Parseur sits behind the tools your HR, finance and compliance teams already run. It has no native HRIS or ATS connector, so you connect it through Zapier, Make, Microsoft Power Automate, n8n or Integrately, and that platform lands the extracted fields in BambooHR, Google Sheets, Airtable, your HRIS or an internal database.

1. Build the flow once in the automation platform you already use, mapping each extracted field to its counterpart on the other side.
2. Point the mailbox at it. Forms parsed in Parseur hand their data to that flow the moment extraction finishes.
3. Let the downstream actions fire. An employee file created, a reverification date set, a document filed against the right case.

Once the connection exists, the data moves without a person in the middle, except at the one step where you asked for one.

What it costs to find out

Parseur has a free plan, and paid tiers are priced by document volume. Current tiers sit on the pricing page.

Setup is a mailbox and an email address, which makes the honest test a cheap one. Pull the five worst I-9s in your files, the photocopy of a photocopy and the one shot at an angle on somebody's kitchen table, forward them, and read the fields that come back. If the citizenship checkbox and the expiration date survive those five, the clean ones were never the problem.

Five forms a month or five hundred, Parseur reads them the same way and hands you the fields:

• Every field off I-9s, W-9s, W-4s and I-94s, including handwriting and checkboxes
• A confidence signal on each one, and a review queue for the fields that carry the risk
• An audit trail that keeps the document, the data and the timestamp together

The forms are already signed. The only thing between them and your systems is somebody willing to retype them, and that somebody has better work to do.

Last updated on September 2nd, 2026