Hewto.ai
Products · Paper to 835

Paper EOB to 835 ERA conversion, automated

Hewto.ai automates paper to 835 conversion — turning faxed, scanned, and mailed EOBs and remittance advices into compliant ANSI X12 835 Electronic Remittance Advice files. Balanced, posting-ready remittances for providers, billing companies, and RCM teams — no manual EOB keying.

The pipeline

Paper remittance to 835 in four steps

Hewto.ai handles the full paper to 835 conversion pipeline — from EOB intake to balanced, posting-ready 835 output.

1

Paper remittance in

Faxed, scanned, and mailed EOBs and remittance advices land in Hewto.ai — no format prep required.

2

EOB capture & validate

Vision-language models extract every payment, adjustment, and reason code, with per-field confidence scoring.

3

Paper to 835 conversion

Captured remittance data is converted into a compliant ANSI X12 835 Electronic Remittance Advice.

4

835 out

Balanced, posting-ready 835 files are delivered to your practice management or RCM system via SFTP, API, or webhook.

Agentic Document Extraction

Why Hewto.ai outperforms traditional EOB OCR

We don't use character-by-character OCR. Hewto.ai runs an agentic document extraction pipeline built on vision-language models fine-tuned specifically for EOBs and remittance advices. The model reads the document the way a trained payment poster does — so it captures the adjustments and reason codes a traditional OCR engine silently misses.

Vision-first, not character-first

A VLM sees the EOB as a whole — dense payment tables, adjustment columns, reason code legends, page breaks, and stamps — instead of stitching guesses letter by letter. It tells watermark text apart from value text, label text from value text, and background from foreground, just like a human reader.

Fine-tuned on remittance documents

Our models are fine-tuned on real EOBs across payers and layouts. They know a CAS adjustment row is a payment reduction with a reason — not just characters in a column — so extractions map cleanly to 835 segments, not raw text.

Agentic reasoning, per field

Each payment and adjustment runs through a purpose-built agent that verifies context, cross-checks claim totals against service lines, and self-corrects when the math doesn't balance — judgment a deterministic OCR pipeline can't make.

Traditional OCR misses
  • Adjustment columns misaligned with the wrong service line
  • Reason codes separated from the adjustments they explain
  • Multi-page EOBs split into orphaned claims
  • Check totals that never reconcile to claim payments
  • Watermarks and background print bleeding into amounts
  • Label text confused with the value text it describes
Hewto.ai captures
  • Every adjustment aligned to its claim and service line
  • CARC / RARC reason codes attached to the right CAS segment
  • Multi-page remittances stitched into complete claims
  • Payments that balance — line items, claims, and check total
  • Watermark text separated from real value text — background noise ignored
  • Label text distinguished from the value text it describes
One-click VLM fine-tuning

Fine-tune a vision-language model at the click of a button

Hewto.ai provides vision-language model fine-tuning at the click of a button. Upload a handful of labeled samples of EOB and remittance forms — or any new form variant, payer template, or document type you handle — and Hewto.ai fine-tunes a purpose-built VLM for it. No ML team, no training pipelines, no infrastructure to manage.

  • Upload samples, click fine-tune, ship
  • No ML engineers or training pipelines required
  • Custom accuracy on your document mix
  • Ready to run in minutes, not weeks
Talk to us about fine-tuning
Fine-tune workflow
  1. 01Upload labeled samplesDrop in 20–200 examples of your target form.
  2. 02Click Fine-tuneHewto.ai kicks off a fine-tuning run — no config, no code.
  3. 03Ship the modelYour custom VLM goes live behind the same API endpoint.
Full remittance coverage

Every section of the paper EOB

Hewto.ai captures every section of the remittance document — so nothing is re-keyed by hand during payment posting.

  • Payer, payee, and check / EFT payment details
  • Claim-level paid amounts, patient responsibility, and status
  • Service-line payments, units, and procedure codes
  • CAS adjustment groups with CARC / RARC reason codes
  • Provider-level adjustments and interest (PLB)
  • Multi-page and multi-claim remittances, stitched and balanced
remittance.835 (X12 835)
ST*835*0001*005010X221A1~
BPR*I*450.00*C*ACH*CCP~
TRN*1*CHK987654*1877260001~
NM1*PR*2*UNITEDHEALTHCARE*****PI*87726~
NM1*PE*1*SMITH*JOHN****XX*1234567893~
CLP*PATCTRL001*1*500*350*150*MB~
SVC*HC:99213*150*100**1~
CAS*CO*45*50~
CAS*PR*2*100~
PLB*1234567893*20250930*L6*0~
SE*11*0001~

Generated automatically from a scanned EOB — paper remittance to 835 ERA.

Paper EOB → X12 835 loop / segment mapping

  • Payer & payee details → NM1*PR / NM1*PE (Loops 1000A / 1000B)
  • Payment amount & method → BPR segment (check or ACH/EFT)
  • Check / EFT trace number → TRN segment
  • Claim totals & status → CLP segment (Loop 2100)
  • Service-line payments → SVC + DTM segments
  • Adjustments & reason codes (CO / PR / OA) → CAS segments
  • Provider balancing adjustments → PLB segment

Why teams switch to Hewto.ai

Post paper remittances automatically — no manual EOB keying
Balanced output: BPR totals, CLP sums, and service lines always reconcile
CAS adjustments preserved with CARC / RARC codes for denial workflows
Handles any payer EOB layout — no templates to build or maintain
Full audit trail from paper EOB image to 835 transaction
SFTP, API, or webhook delivery into your PM / RCM system
EOB capture & 835 generation

Remittance data extraction built for payment posting

EOB OCR at any quality

Reads scanned, faxed, and photographed remittance documents — dense tables, multi-page EOBs, and stamped or annotated copies.

Payment balancing built in

Every 835 is validated so claim payments sum to service lines and the BPR payment matches — catching extraction errors before posting.

ERA 835 generation service

Full-service paper remittance to 835 conversion with human-in-the-loop QA and SLAs — an extension of your payment posting team.

835 conversion API

POST a paper EOB image or PDF and receive a compliant X12 835 in response. Webhooks, sandbox credentials, and predictable schemas.

Paper to 835 FAQ

EOB to 835 conversion, answered

Common questions about Hewto.ai's paper to 835 conversion service for EOBs, EOPs, and paper remittance advices.

Paper to 835 conversion turns paper remittance documents — EOBs (Explanations of Benefits), EOPs, and paper remittance advices — into compliant ANSI X12 835 Electronic Remittance Advice files. Hewto.ai automates the entire workflow: EOB capture, payment data validation, balancing, and 835 EDI generation.

Hewto.ai ingests the paper remittance from fax, scan, digital mailroom, or SFTP, uses vision-language models fine-tuned on EOB layouts to capture every payment, adjustment, and reason code, validates and balances the data, and maps it to a compliant 835 transaction — delivered to your practice management or RCM system with no manual keying.

Any payer remittance document: EOBs, EOPs (Explanations of Payment), paper remittance advices, and Medicare SPRs — from any payer, in any layout. Because Hewto.ai's models are vision-based rather than template-based, layout variations across payers are handled automatically, and the models can be fine-tuned on your specific EOB formats.

Every adjustment on the paper EOB is captured with its group code (CO, PR, OA, PI, CR) and CARC / RARC reason codes, then written to CAS segments at the claim or service-line level exactly as the payer adjudicated it — preserving the data denial-management and posting workflows depend on.

Every generated 835 is validated for financial balance: service-line payments (SVC) plus adjustments (CAS) equal the claim charge, claim payments (CLP) sum to the total payment (BPR), and provider-level adjustments (PLB) reconcile the remainder. Files that don't balance are flagged for review instead of posting bad data.

Providers, billing companies, RCM vendors, and payment posting teams buried in paper EOBs — especially where payers still mail paper remittances. Converting paper to 835 lets those payments auto-post in the practice management system instead of being keyed by hand.

Yes. Multi-page remittances are stitched together, claims are split into separate CLP loops, and check-level totals are reconciled across every page — a common failure point for template-based EOB OCR that Hewto.ai handles natively.

Hewto.ai achieves 99%+ field-level extraction accuracy on remittance documents, with per-field confidence scores and optional human-in-the-loop review so low-confidence payments and adjustments are verified before they post.

Yes. Remittance documents contain PHI and are processed in HIPAA-aligned environments with encryption in transit and at rest. BAAs are available for covered entities and business associates, with hosted API, private cloud, and on-premise deployment options.

Send us a sample batch of your paper EOBs — including your densest multi-page remittances — and we will convert them to 835 and share a field-level accuracy and balancing report so you can compare Hewto.ai against your current keying or OCR process before signing anything.

Ready to see Hewto.ai on your documents?

Book a personalized demo and watch us process your HCFA, UB-04, and dental claims live.