A product by

Bulk upload

Drag in the folder, and every document lands in the right file

Payslips, contracts, mandatory notices: hundreds of PDFs to sort by hand become one batch to check in four steps. You confirm, Anagrafika files them.

Request a demo Anagrafika overview

The file name does the sorting

Whoever manages staff for several companies receives documents in waves: the monthly payslip, hiring notices, signed contracts. Opening them one by one to see whose they are and where they go is the job Anagrafika takes off your hands. It reads surname, first name, employee ID and document type from the file name, checks them against the chosen company’s records and shows you only what does not match. Every correction you make is remembered: the next batch runs on its own. And with the “AI document analysis” extension it recognises employee and type even when the file name says nothing.

How it works

  1. You upload the batch

    Pick the company and drag in files or whole folders: up to 300 documents per batch, 50 MB each. The preview tells you straight away how many are ready and how many need fixing; identical files are ignored.

  2. You fix only the exceptions

    In “Employee mapping” you see who is “OK”, whose name differs and who was not found: keep the record, correct it, remap to another employee or move the files to the bin. In “Document mapping” you confirm the suggested types.

  3. You confirm and import

    The “Summary” tells you how many documents will be created and for how many employees. “Confirm and import”, and the files go into the employee files. The batch stays in the “Upload list” with its history and a downloadable report.

Three hundred documents, four steps, and you only check what does not add up.

What you can do

  • Recognition from the file name

    The ideal format is Surname-Name-EmployeeID-Type.pdf, but Anagrafika also understands “Mario Rossi – Hiring notice” and files that carry only the type. Tails such as “- signed” or “(1)” are ignored.

  • Learns from your corrections

    When you pick a file’s type by hand, Anagrafika remembers it: from then on, documents with the same name match on their own, including those already in the batch. The first upload teaches, the next ones flow.

  • No duplicates

    Identical files uploaded twice are ignored on arrival. A document the employee already has in their file, with the same name and type, is flagged as a duplicate and skipped.

  • The bin, for files to deal with later

    An employee who does not match does not block the batch: move them to the bin, or use “Proceed with the ready ones”, and import the rest. From the “Upload list” you resume each file by choosing employee and category, and it goes straight into the file.

  • History and XLSX report

    Every batch has author, date, company and status, plus a history of what happened: remaps, corrections, skipped duplicates. “Download XLSX report” gives you the outcome of each file, with employee, category and type.

  • “AI document analysis” extension

    When file names do not help, the AI reads the content — text, OCR of scans, the page image — and recognises the employee from tax code, employee ID or name. It suggests the type; confirming it stays with you.

  • Documents stay in-house

    The analysis runs on Kamzan’s local AI models: employee documents are not sent to outside services and the extracts are deleted when the batch ends. It works in the background: close the window, and administrators are notified when it is ready.

Frequently asked questions

How should I name the files so they are recognised?

The format that works best is Surname-Name-EmployeeID-Type.pdf, for example “Ferrari-Giulia-00123-Employment contract.pdf”: with the ID the employee is found immediately. Without an ID, Anagrafika searches by name and asks you to confirm; if the file name contains only the type, you assign the employee in the next step.

Can I upload documents from different companies in the same batch?

No: one batch, one company. Employees are searched only within the company chosen at the start, so a namesake from another company never ends up in the wrong file. If many files find no match, Anagrafika warns you that you may have picked the wrong company.

Does the AI analysis send employee documents to an outside service?

No. The “AI document analysis” extension uses models that run inside the Kamzan infrastructure: payslips, contracts and ID documents are not sent to outside AI providers. The extracted text is used only during the batch and removed at the end.

Who can use the AI analysis, and does the AI import on its own?

The extension is off by default and is switched on for your company on request. Once active, Administrators use it; an administrator can grant it to other user types. The AI imports nothing on its own: it pre-matches employees and types, then the mapping steps stay the same and a person always confirms. If a document tells it nothing, the match from the file name still holds.

What happens to the files I move to the bin?

They are not imported, but not lost either: they stay on hold in the batch detail, in the “Upload list”. Once you know whose they are, you resume them one by one by choosing employee and document category, and they go straight into the file.

The other Anagrafika features

Bring order to your personnel files

Request a demo: we’ll show you how Anagrafika collects, organizes and keeps personnel files complete.

Request a demo