Mandates & files — 3,000 claim filings, structured and verified
How a major proceeding under manual data capture and time pressure produced a new product — Intelligent Intake.
In production
work. ERLEDIGT.
Who is Puttinger.Vogl?
Puttinger.Vogl Attorneys is a law firm from Upper Austria specialised in insolvency law. The firm takes on tasks as insolvency administrator — handling insolvency proceedings is part of its daily practice. That means, among other things: managing the insolvency estate, reviewing and registering creditor claims, negotiating with parties involved, and the legal preparation of realisation.
An insolvency proceeding is complex, time-critical and requires absolute accuracy. Especially in larger insolvency cases: every claim has to be reviewed formally and factually. Every registration has to be documented, categorised and prepared for further legal processing. In a large proceeding there can be thousands of cases — in parallel, under constant time pressure, with deadlines that don't wait.
The problem
One concrete large proceeding: over 3,000 claim filings in a single insolvency case. The filings arrived through the most varied channels — as emails, as PDF invoices, as free-form text, partly handwritten notes, bank statements, delivery confirmations. Every source, every format, no standards.
The problem: Advocat, the law-firm software used as the central case-management system, could not process this unstructured data. The workflow was manual: capture every claim individually, read out the relevant data, bring it into a structured form, review it formally and factually, categorise it, and only then — enter it into the system. Every single one of the 3,000 claims. Under extreme time pressure. With deadlines that keep running.
That tied up staff who were needed for the core legal work. Other attorneys couldn't wait until the data was finally structured. The delay wasn't just inefficient — it was a legal risk.
Work in — result out
Input
Unstructured claim filings — emails, PDFs, free-form text, invoices, bank statements. Over 3,000 in a single proceeding. Each one different.
UNOY works
- • AI extracts parties, amounts, periods from unstructured documents
- • Automatically assigns claims to their category
- • Checks formally against the rulebook
- • Identifies gaps and inconsistencies
- • Human-in-the-loop: every claim is reviewed
- • Transfers structured data directly into Advocat
Result
Fully structured claim filings, formally checked, categorised and ready for further legal processing.
- ✓ Inside the Advocat system
- ✓ On deadline
- ✓ Auditable & documented
- ✓ No gaps
3,000+
claim filings in a single proceeding
Advocat-Integration
Direct API handover of structured data
Human-in-the-Loop
Every claim reviewed before handover
→ Intelligent Intake
A new product emerged from this case
“With UNOY we were able to handle a major proceeding digitally and efficiently. Claim filings were captured automatically, reviewed and transferred straight into Advocat. That accelerated our workflows and created the basis for a new product.”
Supervision model
Self
You build your own digital twin that delivers results. Every rule defined by you, every result under your control.
Your digital expert — built by you.
Partner
UNOY's expertise is included. You work together with the system — we set it up, you keep control.
Expert knowledge included — we set it up, you keep control.
Supervised
The entire work is delivered as a finished result — with legal accountability through Skribe Attorneys.
You hand off the case — we deliver the finished result.
Why UNOY?
The Advocat law-firm software alone couldn't solve the problem — the missing interface for unstructured inputs was a bottleneck. Manual data capture under time pressure was impossible and too error-prone. UNOY took the decisive step: the intake — the moment when unstructured work comes in and structured, auditable data comes out.
The workflow in UNOY is configurable — not for programmers, but for lawyers. Every review rule is documented, every decision traceable. Human-in-the-loop doesn't mean black box: the attorney sees what the AI has extracted, reviews it, and only then it moves on. For insolvency proceedings that's not optional — it's mandatory.
At the same time, this case was so practical that "Intelligent Intake" emerged from it as a standalone product. A product that solves the same problem for other companies — the leap from unstructured inputs to structured, verified data. Deterministic AI, not generative. Governance built in, not optional.
What we learned
The Puttinger.Vogl case showed us three things we had previously underestimated.
First: the real problem sits upstream of the legal work. The biggest time sink wasn't the legal assessment of a claim — it was assembling, deciphering and structuring the incoming data. Emails without a clear subject line, PDFs with contradictory details, handwritten additions. The lawyers spent most of their time turning chaos into order before they could even start working legally.
Second: human-in-the-loop isn't a compromise — it's the decisive advantage. The AI extracts faster and more completely than any human. But humans see connections that no model recognises. The combination of the two — machine speed and human judgment — was the reason the proceeding could be completed on time.
Third: pressure produces product. This case wasn't an innovation project. It was a necessity. 3,000 claims, tight deadlines, no alternative. Exactly that pressure led us to build Intelligent Intake not as a planned feature — but as an answer to a real problem. You can tell from the product: it solves exactly what it has to solve. No more, no less.
Technical deep-dive: how we built it with UNOY
expand_more
AI-based data extraction from unstructured documents
Emails, PDFs, free-form text, invoices — automated OCR and NLP for data capture
Automatic mapping to claim logic and categorisation
Parties, amounts, periods are correctly recognised and classified
Formal legal review against the rulebook
Predefined review criteria, documented decision logic
Human-in-the-loop control step
Domain expert validates every step before handover to Advocat
Advocat-API-Integration
Direct, structured data handover into the case-management software
Deadline and status monitoring
Automatic tracking of processing steps and deadlines
Audit trail for every claim
Complete documentation from input to handover, traceable and auditable
Scales to thousands of claims
Parallel processing at constant data quality
More customer cases
GL Inkasso
Thousands of debt-collection cases — CSV to Advocat with automated dunning letters
WU Legal Tech Center
Build regulatory workflows — in the Legal Tech certificate programme
SKW Schwarz
Review engagement letters, manage playbooks, sign off review tasks
3,000 claim filings. Structured. Hand us your cases.
30 minutes. Your law-firm case. No slide deck.