Cloud Edition See On-Premise Edition →

Natalia Analytics security Hosting, access, data retention

Reference for CISOs and DPOs: data location, isolation between organisations, access and roles, retention and purge, subprocessing.

Stored in France (Paris) Isolation between organisations Retention set per organisation Daily automatic purge

Data location

Conversation data (audio recordings, transcripts, analyses) is stored in France, in Paris; where sub-processors process data is listed in the DPA. Transcription and analysis rely on a GDPR-compliant AI engine. Exchanges between your browser and Natalia are encrypted (HTTPS).

Isolation between organisations

Every piece of data belongs to one organisation. Every read first checks that the user is a member of that organisation; every change also checks the permission it requires. A non-member receives the same refusal whether the organisation exists or not.

A group reads its agencies; an agency reads neither its siblings nor its group. The functional side is covered on the multi-tenant management page.

Access and roles

Each user signs in with a personal account. The API is used with a user's session, never with a shared key: every call carries the rights of its author.

Two roles per organisation: Administrator and Member. Every member reads the conversations, their recordings and Trends; only administrators change them. Details on the roles section.

Data retention

Each organisation has its own conversation retention period, in days, set according to your contract.

  • Daily purge. Every day, a scheduled purge deletes the conversations past the period: the audio recording first, then the transcript, the analysis and the score corrections.
  • Summaries. Category summaries outlive the purge of their sources and store no transcript excerpt.
  • Deletion. A deleted conversation becomes inaccessible immediately; the retention purge erases it permanently. With no period set, permanent erasure is carried out on request, under the DPA.

A deletion request from a data subject is handled under the terms of your contract (DPA). Exporting and deleting an organisation are covered on the reversibility page.

Subprocessing

The list of subprocessors (hosting, transcription, AI engine) and the commitments that come with it are set by your contract (GDPR art. 28 DPA). The up-to-date list is available on request at [email protected].

Breach notification and vulnerabilities

In the event of a personal data breach, Natalia notifies your organisation under GDPR article 33 and within the time set by your contract. Contact: [email protected].

Vulnerability disclosure: see getnatalia.com/.well-known/security.txt.

Data retention, frequently asked questions

How long does Natalia keep conversations?
For the retention period set for your organisation, in days, according to your contract. A daily purge deletes the conversations past that period. A deleted conversation becomes inaccessible immediately; the retention purge erases it permanently. With no period set, permanent erasure is carried out on request, under the DPA.

What does the purge delete?
The whole conversation: its audio recording first, then its transcript, its analysis and its score corrections.

Do summaries keep data from purged conversations?
Category summaries outlive the purge of their sources, but store no transcript excerpt: a purged source shows as purged.

Where is the data hosted?
Stored in France, in Paris. Where sub-processors process data is listed in the DPA.