Data protection & security
Last updated: 9 July 2026
Security and data isolation are foundations of SellFactory. This page describes measures that are implemented in the platform today. It uses only verifiable claims; where something is planned rather than active, it says so.
Tenant isolation
SellFactory is multi-tenant. Data belonging to one customer (tenant) is isolated from every other tenant at the database layer using PostgreSQL Row-Level Security (RLS). Isolation policies are enforced at runtime through a dedicated database role that is not the schema owner and cannot bypass those policies. Application services connect as this restricted role; queries without a valid tenant context return no tenant-scoped data.
Credential protection
Credentials for connected accounts are encrypted at rest using AES-256-GCM and are never stored, logged or returned in plaintext. Encryption keys are held in the server environment, separate from the application code and from version control.
Authentication & access
- Database-backed sessions: session tokens are stored only as salted hashes, never in plaintext.
- Passwords are hashed with a modern memory-hard algorithm (Argon2id).
- Role-based access control governs what each user can see and do within a tenant.
- Sensitive actions are recorded in an audit log.
Data we do not collect
In its current design, SellFactory works with aggregate commerce and advertising data authorized by the customer. It is designed not to collect end-buyer personal information such as buyer names, shipping addresses or invoices. Marketplace integrations, when live, will request only the minimum data scopes required for the analytics described on this site.
Bring-your-own model
Customers connect their own commerce accounts and their own AI provider. You remain the owner of those relationships, you authorize each integration explicitly, and you can revoke access at any time. Customer data is not used to train AI models.
Hosting
The platform is hosted on infrastructure located in the European Union. Sub-processors are limited to what is necessary to run the service and, for AI features, to the provider chosen by the customer.
Responsible disclosure
If you believe you have found a security vulnerability, please contact us at g.contaldi@cgcomunicazioneglobale.com. We commit to acknowledging legitimate reports and will not pursue legal action for good-faith security research conducted without harming users or data. We do not currently operate a paid bug-bounty program; we do offer recognition.