---
title: "How Intelligent Document Processing Works | AI Fortis"
description: "How intelligent document processing works in ScanSystems: upload or scan, read, extract, verify and archive, step by step with the checks at each stage."
url: https://aifortis.com/en/scansystems/how-it-works/
language: en
last_updated: 2026-11-08
publisher: AI Fortis Ltd. (Sofia, Bulgaria)
---

# How intelligent document processing works in ScanSystems

Documents in, structured data out. ScanSystems captures documents, reads them, extracts the fields you need and files everything in one secure, searchable archive.

- GDPR by design
- Cloud, on-premises or hybrid
- 24/7 operation and support
- A team of 5 in Sofia, EU

Already running in production at organisations in Bulgaria — names under NDA.

[How we prove it →](https://aifortis.com/en/how-we-prove-it/)

From paper to data

## One document, four stations

Every document takes the same path: it is scanned and put in order, read, turned into structured data, and filed where your team can find it again.

1. Scan Pages are scanned fast, turned the right way up and sorted.
2. Read Every page is read automatically, including imperfect scans and photos.
3. Extract Header fields and line items are pulled out, validated and verified.
4. Archive The clean record lands in one secure archive, ready to search and export.

The pipeline

## Capture, read, extract, verify, archive

Every document goes through the same five steps, whether it comes from the scanner in the main office, a phone camera or an upload in a branch office.

1. Capture Scanner, phone, camera or upload. Pages are turned the right way up and kept in order.
2. Read Every page is read automatically, also imperfect scans and photos.
3. Extract Header fields and line items become structured data, including amounts in BGN and EUR.
4. Verify Values are validated and corrected; anything doubtful is shown to a person next to the original.
5. Archive Data and image land together in the encrypted central archive, ready to search.

Three apps

## Scanner, web app and archive, working as one system

The main office scans, branch offices upload, and everything lands in one archive. Each office sees only its own documents.

- Main office Scanner app Drives the scanner, phone or camera, reads and extracts on the spot and files into the archive.
- Branch offices Web app Each office uploads into its own account in the browser; documents are extracted and verified there.
- Everyone with access Archive app The encrypted central archive: browse, search, view, edit and export, with access per user.

### Central archive

Access per office and per userCloud, on-premises or hybrid

## Four steps from paper to data

### 1. Capture

Documents arrive from a scanner, a phone or a camera, or are uploaded in the browser — single pages or multi-page documents. With our proprietary scanning hardware, pages are scanned fast and oriented and sorted correctly before anything else happens.

### 2. Read

Each page is read: the image becomes text the system can work with.

### 3. Extract

The fields you need — header fields such as supplier, number, date and amounts, and the line items — are extracted, validated and corrected automatically.

### 4. Archive

The document image and its structured data are filed in the central archive, where they can be searched, viewed, edited and exported.

## Three apps, one system

### Local Scanner App

The desktop app in your main office. It scans from a scanner, phone or camera, reads and extracts documents on the spot and files them into the central archive.

### Web App

A secure browser app for branch offices. Each office uploads its documents to its own account, where they are extracted and verified, then sent to the main archive with structured data and images.

### Archive App

The encrypted central archive where everything lands. Search, view, edit and export any document, with access controlled per user.

## Why page order and orientation matter

Skewed, upside-down or shuffled pages make every later step harder: reading gets worse, fields end up on the wrong document, and people have to fix things by hand. Our scanning hardware orients and sorts the pages correctly during the scan, so reading and extraction start from clean input.

## A typical day with ScanSystems

In the morning the main office scans the post with the Local Scanner App. Branch offices upload their invoices and receipts through the Web App during the day. Every document is read, its data extracted and checked, and it lands in the central archive. Accounting works from structured data instead of retyping, and anyone with access can find a document by searching for any word in it.

## What to prepare

- The document types you process most (for example supplier invoices) and roughly how many per month.
- The fields you need from each type, and where that data must go next.
- Which offices capture documents, and who in each office should see what.
- Where the system should run: in the cloud, on your own infrastructure, or a hybrid of both.
- Your retention rules: how long documents must be kept.
- A small set of real sample documents for the demo.

## From a branch office to the main archive

Documents reach ScanSystems in two ways. In the main office, the Local Scanner App scans directly from a scanner, a phone or a camera — single pages or multi-page documents — reads and extracts them on the spot, and files them straight into the central archive. In branch offices, staff scan or upload their paper documents in the Web App, each office into its own account. There the documents are extracted and verified, then sent to the main-office archive with both the structured data and the page images. Whatever the route, everything lands in one encrypted archive where your team can search, view, edit and export it.

## Frequently asked questions

### Can we capture documents with a phone?

Yes. The Local Scanner App captures documents from a scanner, a phone or a camera.

### Does ScanSystems handle multi-page documents?

Yes. Single pages and multi-page documents are both supported.

### What does a branch office need?

A browser. Branch offices upload documents through the Web App to their own account.

### What does the proprietary hardware do differently?

It scans fast and orients and sorts the pages correctly during the scan, so reading and extraction start from clean pages.

### Do people still check the extracted data?

Fields are validated and corrected automatically, and documents uploaded by branch offices are verified before they go to the main archive. Data in the archive can still be edited.

### Does every office need the Local Scanner App?

No. The Local Scanner App is for the main office; branch offices work in the browser through the Web App.

## Ask about ScanSystems

Bring a stack of your documents — we'll show ScanSystems on them.

- GDPR by design
- Cloud, on-premises or hybrid
- 24/7 operation and support
- A team of 5 in Sofia, EU

Already running in production at organisations in Bulgaria — names under NDA.

[How we prove it →](https://aifortis.com/en/how-we-prove-it/)

---
Source: https://aifortis.com/en/scansystems/how-it-works/
