Anonymize real estate documents for AI
To anonymize real estate documents for AI, you start from the files an agency or a property manager (syndic, the manager of a French co-owned building) handles every day: a lease in Word, a check-in inventory in PDF, a management or sales mandate, general meeting minutes, a rental management spreadsheet in Excel. Each one names a tenant, a landlord, a co-owner, and carries identifiers that point to a home unambiguously: lot number, lease number, meter number. Biffer recognizes them on your computer, with no connection, replaces them with stable aliases and highlights them in black in the output file.
On this page we go through these files one by one, how to redact them, the framework of the French Hoguet and ALUR laws, and the AI uses that follow.
Real estate files and what they contain that is sensitive
Sales, rental management and co-ownership produce different files that speak about the same people and the same properties. Here are the ones we find everywhere.
Leases and amendments (Word, PDF)
Lease number, identity of the tenant and the guarantor, address of the home, rent, lot number.
Check-in and check-out inventories (PDF)
Names of the parties, meter readings with electricity (PDL) and gas (PCE) meter numbers, remarks on the condition of the home.
Management and sales mandates (Word)
Mandate number, identity of the owner, negotiated fees, IBAN for rent transfers.
Tenant application files (PDF)
ID document, payslips, tax notice, employment contract: the most sensitive file in the profession.
General meeting minutes and calls for funds (PDF, Excel)
Co-ownership number, names of the co-owners and their shares, named votes, unpaid charges cited in the meeting.
Rental management spreadsheets (Excel, CSV)
One row per lot: tenant number, rent, arrears, rent receipts, plot.
Text PDFs, Word, Excel, CSV and plain text go through the same process (see redact a PDF); images and scanned PDFs go through built-in optical character recognition, offline, then are redacted by blacking out the pixels, with no possible restoration.
Redacting these files in practice: formats, aliases, black highlighting
Everything happens on your computer, with no connection (see offline software).
- You drop the file with the real estate profile enabled. The local analysis highlights people, addresses, dates, IBANs, lot, lease, mandate and meter numbers.
- You untick false positives (the name of a residence, a street cited as a landmark) and add what is missing, all occurrences at once.
- Each item becomes a stable alias (LOCATAIRE_001 for a tenant, LOT_003, BAIL_001 for a lease; the alias labels are in French), identical from one file to the next for the same property.
- In Word and PDF, redacted data is highlighted in black, an option enabled by default: you see at a glance what was removed. In Excel and CSV, columns and formulas are preserved.
- You choose whether to save the mapping table, encrypted (Argon2id and AES-256-GCM) or in plain text. With it, the real names are restored in the AI's answer; without it, the redaction is irreversible.
The queue takes a whole building, in reviewed or automatic mode. Biffer is free for five files per month, all redaction features included, after a one-time email confirmation in the software, with no password. Beyond that, Biffer for Mac or Biffer for PC costs €10 excl. VAT (€12 incl. VAT) per computer, paid once by card via Stripe: a lifetime license for the version purchased, tied to the computer, unlimited files, each major version sold separately. Every feature is included from the free plan (see pricing).
The real estate profile adds property and contract identifiers to the general detections. Here is the exact list from the software:
| Field | Where you find it |
|---|---|
| Lot number | Leases, co-ownership bylaws, calls for funds |
| Lease number | Leases, rent receipts, letters to the tenant |
| Mandate number | Management and sales mandates, mandate register |
| Co-ownership number | Meeting minutes, calls for funds, pre-sale statements |
| Rent receipt number | Rent receipts, account statements |
| Meter number (PDL / PCE) | Inventories, readings, energy contracts |
| Plot number | Title deeds, planning documents, new-build programs |
| Building permit number | New-build programs, off-plan sales, planning files |
| Tenant number | Rental management software, exports, reminders |
On top of these come the fields shared by all professions (file number, reference, contract number, employee number) and the general detections: people, emails, phone numbers, addresses, French company numbers (SIREN and SIRET), French social security number, IBAN, license plates, dates. An identifier specific to your agency can be added by keyword (see anonymization by profession).
The Hoguet law, the ALUR law and the GDPR
In France, estate agents, property administrators and syndics are regulated by Law No. 70-9 of 2 January 1970, known as the Hoguet law. Law No. 2014-366 of 24 March 2014 on access to housing and renewed urban planning, known as the ALUR law, tightened the rules on leases and co-ownership and limited the list of documents a landlord may request from an applicant, detailed on service-public.fr (in French).
Two consequences for the profession's files. The application file stays in the agency: tax notices, payslips and ID documents appear there in clear, and only the summary you draw from them, redacted, goes to the AI. The syndic handles the data of every co-owner, votes and arrears included, and the GDPR applies to each document. Biffer prepares, you validate: the property manager reviews the preview and has the final word on what is shared.
AI use cases once the files are redacted
Redacted meeting minutes keep their resolutions and votes: LOT_012 votes against resolution 7, the arrears of LOCATAIRE_004 are discussed. The AI can use them without knowing anyone.
- Summarize thirty pages of general meeting minutes.
- Draft a reply to a tenant complaint from the lease and the history.
- Compare the check-in and check-out inventories before returning the deposit.
- Extract the special clauses from a series of leases.
The redacted file can be pasted into ChatGPT, Claude or Gemini; the page features details each step. The answer, loaded back into Biffer with the mapping table, gets the real names back, still locally. Biffer is designed in Lyon by Romain Rissoan, its founder, trainer and consultant, at OPCADIA.
Frequently asked questions
Should a tenant application file be redacted or kept in the agency?
Keeping it in the agency remains the rule: a tax notice or an ID document belongs there, and these documents, often scanned images, call for permanent redaction by blacking out rather than pseudonymization. What is useful to redact is the summary you draw from them: an Excel sheet of applications with income, contract type and guarantor, where Biffer replaces the names with CANDIDAT_001, CANDIDAT_002 and keeps the amounts. The AI can then rank the files on objective criteria, and the mapping table gives you the names back on your computer.
Is a PDL or PCE meter number really data to redact?
Yes. The electricity delivery point (PDL) and the gas metering point (PCE) are attached to a specific home and allow the network operator to find its address and occupant. In an inventory, they come with the readings and the names of the parties: together, they identify a person. Biffer's real estate profile recognizes them and replaces them with a stable alias (COMPTEUR_001), identical at check-in and check-out, which lets the AI compare both inventories without knowing the home.
How do you redact all the files of a building before a portfolio analysis?
You drop the leases, inventories, minutes and management spreadsheet into the queue, then choose reviewed or automatic mode. Aliases are shared across the files in the batch: lot LOT_003 and its tenant LOCATAIRE_003 carry the same alias in the lease, the rent receipt and the Excel sheet. The AI can then cross-reference late payments, approved works and lease expiries, with every row protected from re-identification. For a first building, we recommend reviewed mode.
Your business documents anonymized, ready to be published or handed to an AI tool.
Biffer runs on your computer without sending a single file. Free for five files a month, unlimited for €10 excl. VAT per computer.
100% offline · Free: 5 files a month · No file sent