Characters to avoid in file and folder names
File names affect URLs, compatibility between systems and how efficiently you work. A small detail that prevents many problems.
Read the article
For lawyers, court experts, auditors and companies
From archive to queryable evidence.
A modular ecosystem that digitises, processes, analyses and secures your case files: in an isolated environment, under your organisation’s own keys.
- Ingestion
- Preparation
- Processing
- Analysis
- Storage
- 5Sequential tiers, available separately or together
- 450–600 dpiScanning resolution, adapted to the condition of the material
- AES-256Encryption at rest, with your organisation’s keys
Where the problem starts
The same obstacles, year after year
Legal and financial professionals face the same obstacles, year after year:
“We lost a case in court because a document was not identified in time. The client no longer even remembered it existed.”
- 01Massive volumes of unstructured documents, in many formats.
- 02Difficulty organising and retrieving information, with a real risk of missing essential evidence.
- 03Critical details overlooked in case files or in complex financial statements.
- 04Extended court timelines, with the inevitable re-reading and re-learning of the file.
- 05High costs and risks generated by manual processing.
- The archive stays a static repository, precisely when it should be the source of the argument.
What we do
SCANALYTIX.AI turns the archive into an interactive intelligence asset.
We take documents in whatever state they are (boxes, binders, digital files, media archives) and return them as a structured, numbered, indexed and queryable case file, with every statement traceable to the page it came from.
The service covers the entire life cycle of the case file and is delivered in five sequential stages, each preparing the ground for the next.
See the servicesThe five tiers
Tiers can be requested separately or together.
- 0
Ingestion
MICAREXPUFThe raw archive, physical or digital, becomes an ordered and verified digital stream.
- 1
Preparation
SARAHNot archiving but digital curation: double inventory, quality score, 450–600 dpi.
- 2
Processing
ZARStatic documents become structured entities: BATES, validated OCR, indexing.
- 3
Analysis
ZARIAFrom manual search to a strategic dialogue with the case file, EDRM-compliant.
- 4
Storage
REBEFREBetween hearings: low cost, maximum security, instant reactivation.
Two delivery models
The choice is yours.
Your documents can remain entirely on your premises, or be collected by us. These are two distinct models, with different legal implications. The model is agreed before contracting and recorded in the contract.
| Capabilities | Model A: Processing on the Client’s premises | Model B: Full service, at SCANALYTIX.AI premises |
|---|---|---|
| Where the documents are | They never leave your premises, except as specified under “Data transfer”. | They are collected, transported and processed at SCANALYTIX.AI premises. |
| Dedicated IDP equipment | Installed at your site, under rental or operating lease. | Equipment and infrastructure belong to SCANALYTIX.AI. |
| Who handles the documents | Your staff, trained by us, or SCANALYTIX.AI staff seconded to your premises. | Vetted SCANALYTIX.AI staff, under NDA and a documented chain of custody. |
| Data transfer | No document transfer takes place. Processing runs on your infrastructure. Exception: if you opt for interactive dialogue with an external language model (Tier 3), the content used in the dialogue is transmitted to the provider named in the contract, on EU infrastructure. Exposure is reduced through pseudonymisation (Tier 2). | Transfer takes place. The custody procedure in Annex 1 applies in full. |
| Suited to | Files under strict professional secrecy, recurring volumes, organisations with strict internal document-security policies. | Organisations that do not wish to manage equipment and staff and prefer a turnkey service. |
| Legal framework | Data processing agreement (Annex 5), signed before the first document. | Data processing agreement (Annex 5) plus the custody procedure (Annex 1), signed before collection. |
One rule, valid in both models: no document enters processing before the data processing agreement is signed. This includes the archive assessment stage and any demonstration on real documents.
Why us?
SCANALYTIX.AI ≠ generic software
Functional automation, not fictional.
- 01
We process the physical archive, not just the PDF: from the box to a BATES-numbered case file.
- 02
The deterministic layer (inventories, validated OCR, dtSearch search, rule-based extraction) sits beneath the AI layer, so every answer can be checked at page level.
- 03
The team includes lawyers, judicial experts and information-security specialists who calibrate what is searched for in each case.
What you get
Compliant evidence, time recovered, documented diligence.
Legal integrity
Every piece of evidence extracted is compliant, traceable and deliverable to the court, with BATES numbering and in line with the EDRM model.
Efficiency
Analysis time drops from weeks to hours, through deep search across any document type and direct dialogue with the case file.
Safety
Isolated environment, equipment on your premises (Model A), encryption with your organisation’s keys, complete audit trail.
Expertise in the loop
Lawyers, court experts, financial auditors and information-security specialists, in the same project team.
A complete legal framework
Data processing agreement, documented chain of custody, deletion and destruction certificates. Outsourced processing takes place within a documented framework that lets you demonstrate due diligence towards your client and the supervisory authority.
The first step
The archive assessment
A report describing what you have, what can be processed, how long it takes and with what risks: the document on which we commit to a firm deadline and volume.
Request the assessmentHow we work together
Three steps, in this order.
No exceptions and no shortcuts.
- 01
Signing the processing framework
The first document signed is the Data Processing Agreement (Annex 5). We do not collect or assess any real document before it is signed. The rule also applies to any demonstration on real documents.
- 02
The archive assessment
We assess the archive on site: the physical condition of the material, the degree of standardisation, the real volume, the document and media types. The result is a report describing what you have, what can be processed, how long it takes and with what risks. The assessment is a service in itself and is invoiced separately.
- 03
The five tiers
Tiers 0–4 can be requested separately or together. Each project is assessed and configured according to its complexity and the Client’s specific requests.
Our commitment: we assess the archive, then commit to a firm deadline and volume. SCANALYTIX.AI’s technical capacity is not the limiting factor; the physical condition of the material is.
Who we are
Legal-tech, built by lawyers and experts.
For lawyers, court experts, auditors, financial consultants and corporate legal departments.
The platform processes the volume. Our experts calibrate what is being looked for, to the specifics of your case.
Specialists from six fields, in the same project team
- Legal
- Forensic
- Financial & accounting
- Information technology
- Cybersecurity
- Information security
Technology + partners
The SCANALYTIX.AI EVO ecosystem is built primarily on Canon, I.R.I.S., Alphabet, Anthropic, dtSearch and Tresorit technologies, in an isolated environment.
Providers that actually process data are named as sub-processors in the contract; the others supply equipment or software run locally. Processing takes place on infrastructure in the European Union, and data is never used for training.
The SCANALYTIX Journal
Perspectives on OCR, BATES, advanced search and case-file analysis.
Best Practices for 2026: Intelligent Document Processing
In an increasingly digital legal world, law firms face ever larger volumes of documents. Processing them intelligently, with OCR, AI and machine learning, is becoming a priority for efficiency and competitiveness in 2026.
Read the article
Why advanced search with dtSearch, regex and other tools beats the classic Ctrl + F
Almost all of us use Ctrl + F (or Cmd + F on a Mac) to find a word quickly in a document or on a web page. It is simple, fast and familiar. The problem starts when you work with thousands of documents.
Read the articleMake friends with your documents.
We organise them, you analyse them more easily. The first step is the archive assessment: a clear report on what you have and what can be done.