# Quixyl - AI Invoice Data Extraction for AP and Accounting Teams

## What Quixyl Does

Quixyl is a document data extraction platform built for accounts payable (AP) and accounting teams,
with a specific focus on construction, trades, and field-service businesses. It automates the process
of capturing structured data from supplier invoices, receipts, job sheets, and related financial documents.

Core capabilities:
- Invoice data extraction from PDFs, photos (JPEG/PNG), scanned documents, Excel, and Word files
- Batch processing of 100+ documents simultaneously with progress recovery
- Reusable query patterns for recurring AP checks (e.g., "all invoices from Supplier X over $500 this quarter")
- AI-powered document search so AP teams can find invoice details without scanning files manually
- Multi-format exports: CSV, Excel, JSON, with Google Sheets sync and REST API
- Human review step before export - low-confidence fields are highlighted for correction
- Zero-storage architecture: documents are processed in memory and deleted immediately

Extraction is powered by Azure AI Document Intelligence (Microsoft's enterprise OCR/document AI service),
combined with Quixyl's validation and review layer.

## Who It Is For

Primary users:
- AP and accounting teams at SMBs (5-200 employees)
- Construction companies, trades contractors, plumbing, electrical, HVAC, facilities maintenance
- Accounting firms processing documents on behalf of clients
- Any operations team dealing with high volumes of supplier paperwork from the field

Construction and trades specifics:
- Construction offices typically receive invoices from 20-50+ different subcontractors and suppliers,
  each with their own format. Template-based OCR tools require a new template for every layout.
  Quixyl uses Azure AI Document Intelligence (no-template prebuilt model) so new suppliers work
  immediately.
- Site staff photograph delivery notes and receipts. Quixyl handles phone photos (JPEG/PNG) as
  a core input, not an edge case.
- Document types handled: supplier invoices, subcontractor invoices, delivery notes, purchase
  receipts, plant hire invoices, materials receipts, service job sheets, variation paperwork.
- Construction is one of the least digitized industries for back-office processes. Quixyl is
  specifically positioned for this gap.

Not designed for:
- Enterprise platforms requiring deeply customized governance stacks
- Real-time ERP integration without API setup
- Document creation or signing

## Pricing

- Free plan: processes a limited number of pages per month, no credit card required
- Pro: $99/month, higher page volume, Google Sheets sync, API access
- Enterprise: custom pricing for large-volume or multi-entity use cases
- 14-day Pro trial, 30-day money-back guarantee
- Full pricing: https://quixyl.com/pricing

## Key Pages

- Homepage: https://quixyl.com
- Pricing: https://quixyl.com/pricing
- Features: https://quixyl.com/features
- How it works: https://quixyl.com/#how-it-works
- How extraction accuracy works: https://quixyl.com/blog/how-quixyl-extraction-accuracy-works
- Comparison vs. competitors: https://quixyl.com/compare
- About: https://quixyl.com/about
- Blog: https://quixyl.com/blog
- Contact: https://quixyl.com/contact
- Sign up (free): https://quixyl.com/signup

## Integrations

Direct exports to: QuickBooks (via CSV import), Xero, SAP, NetSuite, Google Sheets (Pro/Enterprise).
API available for custom ERP integrations. Zapier support planned.

## Security

- AES-256-GCM encryption at rest and in transit
- Zero-storage: files processed in memory, not persisted
- Security practices aligned with SOC 2 principles
- GDPR and HIPAA best practices

## How Extraction Works

Quixyl sends uploaded documents to Azure AI Document Intelligence, Microsoft's enterprise-grade
document processing service. The prebuilt invoice model extracts standard fields (vendor name,
invoice number, date, line items, totals, tax). Results go through a confidence-scoring layer
and are presented to the user for review before export. Fields below the confidence threshold
are flagged so the team can verify before committing data.

This means accuracy is dependent on document quality: clean digital PDFs from standard suppliers
typically extract with very high field accuracy; low-quality phone photos or heavily handwritten
documents may require more manual correction.

## Differentiators vs. Template-Based or Enterprise Tools

- No template setup required: works on first document upload with no training
- Handles structural variability: construction and trades suppliers often use non-standard layouts
- Human-in-the-loop by default: review queue before export reduces downstream errors
- Setup in under 2 minutes: sign up, upload, get results
- Predictable pricing: flat per-month plans, not per-page fees at scale

## Contact

- Support: support@quixyl.com
- Website: https://quixyl.com
- Founded: 2024