SCANALYTIX: Case File Processing for Legal and Audit Work
On This Page
SCANALYTIX.AI is a Romanian legal-tech company that turns paper and digital archives into case files that can be searched and cited page by page. It works for lawyers, court experts, auditors, financial consultants and corporate legal departments. It sells a service in five tiers, not a self-service software product.
How this profile came about: Tomer Mozafi requested the listing through the contribute form on 2 November 2026. IDP Software wrote the profile from SCANALYTIX's website. Figures marked as vendor claims have not been verified independently. About the author: Tomer Mozafi.
Company and ownership
The homepage describes SCANALYTIX.AI as a legal-tech company; the footer says "Romania". The privacy page names SCANALYTIX.AI as data controller. The website gives no legal form, address, registration number, founding year, headcount or owner, so none is stated here. The about page tells a founding story: the company began after a court case was lost because a document was not identified in time ("We lost a case in court because a document was not identified in time"). The quote is attributed to a founder, who is not named. The page says lawyers and experts define what the platform searches for, and that one project team combines six fields: legal, forensic, financial and accounting, IT, cybersecurity and information security. It names no team member. The terms of use are governed by Romanian law with the competent Romanian courts. Contact is through tomer.mozafi@scanalytix.ai, +40 722 670 423, the contact page or a 30-minute call booking.
Where SCANALYTIX fits
SCANALYTIX covers the work between a box of files and a lawyer's question: scanning, cleaning, OCR, numbering, indexing and search over the finished case file. It fits legal, forensic and audit projects with large, mixed archives. A team that needs invoice or form capture in a running business process should look at the wider IDP vendor directory.
The service has five tiers that can be ordered separately or together, and two delivery models: processing on the client's premises with dedicated IDP equipment installed on site, or full service at the vendor's premises with collection and transport of documents.
How SCANALYTIX processes documents
| Tier | Name | What the vendor says |
|---|---|---|
| 0 | Ingestion (MICAREXPUF) | The raw archive, physical or digital, becomes an ordered and verified digital stream |
| 1 | Preparation (SARAH) | Double inventory, quality score, 450-600 dpi |
| 2 | Processing (ZAR) | BATES numbering, validated OCR, indexing |
| 3 | Analysis (ZARIA) | Dialogue with the case file, EDRM-compliant |
| 4 | Storage (REBEFRE) | Low-cost storage between hearings, reactivation on demand |
The services page describes OCR with dual-engine validation to find reading errors. Values below a confidence threshold go to an operator for correction. Extraction is either rule-based, "without a generative model", or AI-assisted with the source BATES page shown for each value. Search runs on dtSearch full-text indexing, so the same query on the same index gives the same result. The AI layer is set to answer only from the processed documents. The vendor says the deterministic layer sits beneath the AI layer so that every answer can be checked at page level. No OCR engine, confidence value or model version is named.
Technical specifications
| Feature | Specification |
|---|---|
| Document Types | Physical archives (boxes, binders), digital files, media archives; case files for legal, forensic, audit and financial work |
| Input Formats | Paper scans at 450-600 dpi and digital files; no format list published |
| Output Formats | Numbered (BATES), indexed, searchable case file; pseudonymised copies; answers with page citations |
| Processing Pipeline | Ingestion, preparation, processing (OCR, extraction, indexing), analysis, storage |
| API/Integration | None published |
| Deployment Options | Service on the client's premises with on-site IDP equipment, or at the vendor's premises; hardware by lease or rental |
| Certifications | Procedures "built on the ISO/IEC 27001 structure"; no certificate named |
| Claimed Accuracy | None published |
Deployment, security and pricing
The security page states encryption in transit (TLS 1.2/1.3) and at rest (AES-256), with the archive keys held by the client. External AI models are optional; it lists Claude (Anthropic) and Gemini (Google), processed in the EU and not used for training. The page does not name a hosting provider or location for the vendor's own systems; it refers to an annex of sub-processors. It does not claim an ISO 27001 certificate, so ask for one if you need it. The vendor says a data processing agreement is signed before any document is processed.
The pricing page gives starting prices in EUR, excluding VAT:
| Item | Starting price |
|---|---|
| Case-file assessment | from 200 |
| Tier 0, ingestion | from 0.10 per page |
| Tier 1, preparation | from 0.20 per page |
| Tier 2, processing | from 0.40 per page |
| Tier 3, analysis | from 500 per month |
| Tier 4, storage | from 100 per month per case file |
Other services vary with complexity. IDP hardware is offered on operating lease or rental based on expected annual volume.
Evidence
All results are SCANALYTIX's own claims. The homepage says analysis time "drops from weeks to hours" without data or an example. The image quality score of 94/100 is an example in an interface mockup; no scoring method or threshold is given. "Government-grade security" and "state-of-the-art models" name no standard or model, and no OCR or extraction accuracy is published. No customer, case study or benchmark is named. The site shows logos of Canon, IRIS, Google, Anthropic, dtSearch and Tresorit without saying what the relationship is. Run a pilot on a sample of your own archive.
Resources
Company Information
SCANALYTIX.AI is based in Romania; the website lists no street address. Legal form, registration number, founding year, ownership, funding and headcount are not published on the website, and no company register entry could be confirmed for this brand. The about page lists Canon, I.R.I.S., Alphabet (Google), Anthropic, dtSearch and Tresorit as technology in the SCANALYTIX.AI EVO ecosystem. No LinkedIn page and no RSS or Atom feed were found.