Koncile

Koncile

Koncile · Productivity · Marketing

Koncile is an AI OCR platform for finance, insurance and compliance teams that turns invoices, IDs and statements into structured data and flags fake documents.

Interface preview of Koncile

About Koncile

What Is Koncile

Koncile is an AI document processing platform built around optical character recognition (OCR). It reads scanned and digital files, turns the useful fields into structured data, and then checks that data for inconsistencies, which is the part most OCR tools skip entirely. A finance team drowning in supplier invoices uses it to skip manual re-keying. An insurance desk verifying bank statements uses it to catch problems before payment.

The platform is split into three building blocks: extraction, automation and fraud detection. Extraction handles invoices, IDs, pay stubs, prescriptions and more, with a claimed 99%+ accuracy. Automation splits batches, classifies files, renames them and pushes results into tools like SAP, Excel or n8n. Fraud detection scores every document on a 0 to 1 scale using more than 150 signals.

The catch is positioning. Koncile is aimed at regulated, high-volume operations, and the strongest safeguards sit behind an Enterprise contract. Small teams that just want to read a PDF occasionally will be paying for a lot of machinery they don't need. Teams that batch document extraction daily and want document automation on top get the most out of it. The tool sits in the wider intelligent document processing space, but it leans harder into fraud checks than most products in that group.

Getting Started

  1. Sign up on the Koncile site and start with the free plan, which includes 20 credits to test extraction.
  2. Upload one or more documents and pick a template: Koncile Lite for faster, cheaper extraction, or Koncile Pro for the most complex layouts.
  3. Review the extracted fields and, if needed, adjust your template rules so the right data lands in the right columns.
  4. Connect the output to your stack through the API, a webhook, or an export to SAP, Excel or an automation tool.
  5. Move onto a paid plan once volumes grow, and add fraud detection if document authenticity is part of your workflow.

Product Information

A quick look at Koncile's pricing, supported platforms, and performance.

Free PlanYes
Paid Plans$0 - custom pricing
PlatformWeb, API
DeveloperKoncile
CategoryProductivity · Marketing
Release DateJan 2023
Latest UpdatedOct 2025
Website Visits40.5K
Website Global Rank827.3K
API AvailabilityYes

Best for

The users, tasks, and scenarios where this tool fits best.

Users

  • Finance and accounting teams
  • Insurance and lending operations
  • Compliance and risk staff

Tasks

  • Invoice data extraction at scale
  • Catching duplicate or altered documents
  • Batch document routing

Scenarios

  • Clearing a backlog of documents
  • Onboarding with identity checks
  • Feeding an existing ERP

Key features

High-Accuracy AI OCR

The core extraction engine reads complex layouts, tables, handwriting and skewed scans, with Koncile claiming 99%+ parsing accuracy. It's purpose-built for common business formats rather than generic screenshots. Two templates sit on top: Lite is faster and about half the price of Pro, while Pro handles the messiest documents. That split lets a team trade cost against difficulty per document type, which is more useful than one fixed accuracy tier. Pick Lite for clean files. Reach for Pro when the scan is a mess.

Invoice Line-Item Capture

For invoices, the engine pulls every line item, not just the headline totals. Reference, quantity, unit price and tax all come out as separate fields. A control layer then cross-checks totals and calculations against the captured figures, and that single step catches a class of errors a human reviewer would likely miss under time pressure. If someone quietly edits a unit price, the numbers stop matching and the entry gets flagged. That's the difference between transcription and actually verifying a bill. Good extraction catches typos. This catches edits.

Document Fraud Detection

Every document gets a fraud score on a 0 to 1 scale, computed from more than 150 signals. The scoring blends forensic image analysis, metadata inspection and contextual anomaly detection built with industry specialists. It looks for compression artifacts, altered timestamps, suspicious PDF layers and documents created or modified by AI. It's available on the platform or through the API, but it's Enterprise-only, so smaller accounts can't switch it on. Annoying, but that's how it's sold.

Document Automation Workflows

Beyond reading files, Koncile automates what happens next. So what does that look like in practice? It renames files, routes them and reconciles data with no one touching them. Batches get split and classified automatically, files are renamed to your conventions, and extracted data is reconciled with your reference systems. Confidence scoring sends low-confidence results to a human for review, anomalies get escalated to the right person with context attached, and the whole chain runs without a single manual copy-paste step. The point is to remove the copy-paste step between reading a document and acting on it.

OCR API and Integrations

A REST API and SDKs let developers wire Koncile into an existing pipeline, and the Advanced plan is the entry point for API access. Data exports to SAP, Excel, webhooks, n8n, Make and Power Automate, so the tool slots into systems teams already run. No rebuilding a stack just to read invoices. There's also a dedicated email address that triggers the pipeline on its own, which suits companies that receive invoices as attachments all day.

Compliance and Data Protection

Koncile was founded by a former lawyer and built for regulated environments. The company says it's independently audited under SOC 2 and compliant with GDPR and HIPAA requirements. For teams handling medical or financial records, that paperwork matters as much as extraction quality, because internal review boards ask about it before any rollout. Skip it and the deal stalls.

Pros and cons

Pros

  • Handles messy real-world documents, including handwriting, tables and skewed scans, not just clean digital PDFs. Real scans are ugly.
  • Fraud detection goes past reading text and actually checks whether a document is authentic, which is rare in OCR tools.
  • Modular setup lets teams start with extraction and add automation or fraud scoring later.
  • Integrations with SAP, Excel and automation tools cut the manual handoff between reading and acting.
  • SOC 2, GDPR and HIPAA compliance give regulated buyers a clear answer on data handling.

Cons

  • Pricing is quote-based for real volumes, so budgeting is hard before you talk to sales.
  • Fraud detection is locked to Enterprise contracts, which rules it out for small teams.
  • The free plan's 20 credits run out fast once you're testing more than a handful of documents. That's a taste, not a trial.

Frequently asked questions

It reads business documents with AI OCR, turns the useful fields into structured data, and checks that data for errors or fraud. Think invoices, IDs, bank statements and pay stubs rather than general-purpose PDFs.