Block-based workflow editor
Prompting, testing, versioning, multi-prompt, cost control
Resume line
Retrieval Augmented Generation (RAG)
Roles and permissions
Security, data transparency
Web
Data Transformer
Decision Maker
Extract Maker (with HTML)
Key / Knowledge Manager
Web Creator
SaaS Email, Integration
Cloud privacy managed
On premise
MODULE · Extraction and analysis from documents
Every document becomes structured data, ready to use.
Define the labels you need: amount, supplier, date, order code, any field. ARIdoc pulls them from invoices, contracts, orders, forms and reports, regardless of layout or format. The output is structured and ready for your systems.
Scales by:
Documents processed / year *
·
from €200/month
Guarda ARIdoc in azione
Works on any document
No fixed template that breaks: define the labels and ARIdoc finds them on invoices, contracts and forms from any supplier or layout.
No silent errors
Confidence on every field: below threshold a person steps in before the data enters your system.
Ready for your systems
Structured output: JSON, CSV or a direct API call to ERP and CRM. No manual transformation.
How it works
From raw document to structured data, automatically.
01
Intake
The document enters the process: direct upload, API, email with attachment or a trigger from an external system, ERP, CRM, shared folder. ARIdoc accepts any channel.
02
OCR + AI Extraction
OCR extracts the text and structure. The LLM interprets the context and identifies the fields defined by your labels: even across variable layouts, different formats, multiple suppliers.
03
Normalization
The extracted data is normalized to a predefined data model: same structure, same naming, same format, regardless of where the document came from.
04
Human In The Loop
If the extraction confidence score is below threshold, the process notifies an operator. The operator validates or corrects: then the flow resumes. No uncertain data passes through automatically.
05
Output and integration
The structured data is written to the destination system: SQL database, the ERP's REST API, JSON or CSV file for import. The processed document is archived with its extraction log.
ARIdoc · invoice_2024-0473.pdf
Document
Supplier
Acme Ltd.
VAT No.
IT01234567890
Invoice No.
2024-0473
Date
12/03/2024
Total
€1,240.00
Description
Q1 consulting services
Extracted data
supplier:
Acme Ltd.
99%
vat_no:
IT01234567890
98%
invoice_no:
2024-0473
97%
date:
12/03/2024
96%
total:
€1,240.00
99%
description:
Q1 consulting services
review
Confidence on every field. Below threshold, the field goes to human review before flowing into your system.
Features with ARIflow
From extraction to action, with no manual step.
With ARIflow, ARIdoc becomes the first step of an automated process: the document arrives, the data is extracted, validated and written to your system, all in sequence, with exception handling built in.
Automatic triggers
New document → starts extraction automatically. Via email, shared folder, API or an event from an external system.
End-to-end pipeline
Extraction → normalization → HitL validation → ERP write. A complete workflow, with no manual handoffs between systems.
Exception handling
If extraction fails or confidence is too low, the workflow handles the exception: notifies, queues, retries, without blocking the batch.
Monitoring and SLA
Monitor processing times, error rates and HitL interventions in real time. Automatic alerts if volumes exceed the threshold or SLAs are at risk.
Modules can be purchased individually. ARIflow is optional and expands each module's capabilities.
Pricing
Scales by usage, not by user.
Scales with how many documents you process in a year. No per-user cost, no per-label cost: everything included in the subscription within the fair use clause.
✓ LLM costs included in the subscription
✓ Unlimited labels
✓ Monthly or annual billing
Per 5.000 documents / year
550 €
/month
Tier up to 5.000 · beyond → 800 €/month
No card required
Il tuo volume:
documents / year
Drag or type your volume: the fee is based on the peak of documents / year during the billing period. More than 100.000? Custom quote.
Included in every tier
✓
Customizable extraction fields
✓
Confidence score + Human-in-the-loop
✓
Multiple layouts and suppliers with no hardcoded rules
✓
JSON / CSV / API output
✓
Full traceability of extractions
✓
LLM costs included in the subscription (fair use)
* A "processed document" is a file on which at least one label extraction is run. Re-processing the same document with different labels counts as a new document. Documents over 10 pages are counted as ⌈pages/10⌉ standard documents (e.g. a 30-page contract = 3 documents).
Deployment
Wherever you want, however you want.
Plug and play
SaaS
Hosted by us. Zero infrastructure to manage.
Upon request
On-Premise
Installed on your servers. Data never leaves your perimeter.
Upon request
Air-Gapped
Fully isolated from the network. For high-security environments.
Details on privacy, compliance and architecture → Deploy & Privacy
Bundle discount
The more modules you buy, the less you pay.
Each additional module beyond the first gives you 5% off the total: up to 15% with four modules. Add ARIflow and another 10% kicks in.
The maximum discount is 25%: four modules plus ARIflow.
Example: ARIdb + ARIknow + ARIflow → 15% discount. €23,400/year become €19,890.
ARIchat is included for free and does not count toward the bundle.
1 module
0%
2 modules
5%
3 modules
10%
4 modules
15%
+ ARIflow
+10%
Ready to bring ARIdoc to your company?
Try it on your data · No commitment
Still have doubts?
Book a quick consultationCookies and tracking
We use technical cookies and — with your consent — analytics and marketing cookies to measure traffic and improve your experience. You can accept, refuse or change your choice at any time.
More info in the Cookies Policy →