Blog
Notes on documents, data and automation.
Practical, no jargon. What we learn building LOCRAI and working with the people who process documents every day.
source: quanto-costa-data-entry-manuale.md
category: automation
published: June 25, 2026
read_time: 11m
Manual data entry: how to measure the real cost of your document workflow
A four-variable method to quantify hours, rework and downstream delay — before evaluating any automation tool.
Read the articlesource: automazione-ciclo-passivo.md
category: automation
published: June 24, 2026
read_time: 12m
Accounts payable automation: from invoice to ERP without data entry
Invoices, delivery notes and credit notes: how to automate accounts payable with AI extraction, data verification and ERP integration — without retyping every PDF by hand.
source: dmv-vs-ocr.md
category: underTheHood
published: June 18, 2026
read_time: 13m
OCR, text layer, visual analysis: what really reads a document
A no-frills guide to the different ways of reading a document and when each one is worth it.
source: addio-data-entry-manuale.md
category: automation
published: June 10, 2026
read_time: 12m
Goodbye data entry: what changes when documents read themselves
The hidden cost of re-typing invoices by hand, and what happens to the work when extraction becomes automatic.
source: gdpr-documenti-cloud.md
category: compliance
published: June 2, 2026
read_time: 12m
Documents in the cloud and GDPR: a pragmatic checklist
Retention, data localization, access: what to look at when you entrust your documents to a service.
source: confidence-non-e-una-promessa.md
category: dataQuality
published: May 22, 2026
read_time: 11m
Confidence isn't a promise: how to trust extracted data
A reliability score only makes sense if it comes from objective checks. Here's the difference between saying and proving.
source: integrare-estrazione-nel-gestionale.md
category: integrations
published: May 5, 2026
read_time: 12m
How to connect document extraction to your business system
Export, webhook or API: three ways to get the data where it's needed, without copy-paste.
source: studi-commercialisti-multi-cliente.md
category: useCases
published: April 20, 2026
read_time: 11m
One firm, a hundred clients: managing documents without losing your mind
How multi-organization management changes the day of those who work for many companies at once.
source: software-estrazione-documenti-criteri.md
category: automation
published: November 18, 2025
read_time: 11m
Document extraction software: eight criteria for evaluating a vendor
EU hosting, data verification, API, pricing model: an operational checklist to compare document extraction solutions without trusting made-up statistics.
source: ocr-su-scansioni-qualita-revisione.md
category: dataQuality
published: August 5, 2025
read_time: 11m
OCR on scans: DPI, skew and the human review queue
Scans impose different constraints than digital PDFs. Image quality, stamps, skew: how to set up the workflow and when human review is needed.
source: estrazione-dati-da-fatture-pdf.md
category: automation
published: April 17, 2025
read_time: 12m
Extracting data from invoice PDFs: methods, common errors and how to avoid them
Manual, template, OCR or AI: how invoice PDF data is extracted, where totals and line items fail, and what to ask before automating.
source: intelligent-document-processing-guida.md
category: underTheHood
published: February 6, 2025
read_time: 11m
Intelligent Document Processing (IDP): what it is and when companies need it
IDP is not a synonym for OCR. What an IDP workflow includes, how it differs from simple OCR and RPA, and in which contexts it is worth investing.
source: ocr-e-intelligenza-artificiale.md
category: underTheHood
published: October 28, 2024
read_time: 12m
OCR and artificial intelligence: what each layer does — and when you actually need it
OCR and AI are not the same thing. A guide to the two layers, when text recognition is enough, and when you need a engine that interprets layout and fields.
