Data residency and security
Last reviewed
The short answer
Case360 stores and processes workplace investigation data in Australia. Case records, files, sign-in accounts, backups and the AI features all run on Sydney regions. The only ways case data can leave Australia are two optional features that are off by default and need an admin to switch them on. This page lists exactly where each type of data is handled, including the small amount of technical and billing data handled offshore.
Where your data is stored and processed
| Data | Service | Location |
|---|---|---|
| Case records (cases, people, allegations, interviews, chronology, tasks, findings, audit log) | Google Cloud Firestore | Sydney (australia-southeast1) |
| Files (documents, evidence, recordings, exports) | Google Cloud Storage | Sydney (australia-southeast1) |
| Sign-in accounts and sessions | Supabase Auth | Sydney (ap-southeast-2) |
| Application servers | Vercel | Sydney (syd1) |
| AI text analysis (summaries, Fact Find, allegations, chronology, report drafts) | Anthropic Claude on Amazon Bedrock, Australian inference profile | Australia (ap-southeast-2) |
| AI for audio, video and interview transcription | Google Vertex AI, or Case360’s own Whisper transcription service on Google Cloud Run | Sydney (australia-southeast1) |
| Background jobs (reports, transcription, exports) | Google Cloud Tasks and Cloud Run | Sydney (australia-southeast1) |
| Outbound email (notifications, invitations, external requests) | Amazon SES | Sydney (ap-southeast-2) |
| Database and file backups | Google Cloud | Sydney (australia-southeast1) |
What is handled outside Australia
We would rather tell you exactly than say “everything stays onshore”.
Error monitoring (Sentry, Germany)
When something goes wrong in the app, technical error and performance data is sent to Sentry’s EU data centre in Germany so we can fix it. This can include the signed-in user’s email address, IP address and the page involved. Sign-in cookies and tokens are removed before anything is sent, AI prompts and outputs are never sent, and session replays mask all text, form inputs and media on screen.
Billing (Stripe)
Subscriptions and invoices are processed by Stripe. Card details go directly to Stripe and never touch Case360’s servers. Stripe holds billing contact details only, never case data.
Email-to-case (optional, off by default)
If an organisation’s admin turns on email-to-case, emails forwarded into a case are received by our email provider in the United States before being stored in Sydney. The setting is off unless an admin switches it on, and the admin screen warns about this before it is enabled.
Claude connector (optional, off by default)
If an organisation enables the Claude connector, information a user asks their own Claude account to read is sent to Anthropic under that user’s Claude agreement. It is off by default, needs an admin to acknowledge the data residency change, and can be switched off at any time, which disconnects every user.
Our infrastructure providers (Google, Amazon, Vercel and Supabase) are international companies. Data hosted in their Australian regions stays in Australia, but those providers are also subject to the laws of the countries they are based in.
How Case360 uses AI with your data
- Text analysis runs on Anthropic’s Claude models through Amazon Bedrock’s Australian inference profile, so prompts and responses are processed in Australia.
- Audio, video and transcription run on Google Vertex AI in Sydney, or on Case360’s own Whisper service in Sydney, so recordings do not leave Australia.
- Under Amazon Bedrock’s and Google Vertex AI’s service terms, customer prompts and outputs are not used to train their models.
- Every AI run is recorded and audit-logged, and AI output is labelled in the app until a person has reviewed it.
- Admins control AI use in Admin > AI Controls: which features are on, the minimum role allowed to use them, and organisation and per-person limits.
Access control
- Every organisation’s data is kept separate. Each request is checked against the user’s current membership of that organisation.
- Role-based permissions (Admin, Lead Investigator, Investigator), plus custom roles and teams on Professional and Enterprise.
- Optional restricted case visibility, so investigators only see the cases they are assigned to.
- Two-factor authentication, which admins can make compulsory for the whole organisation.
- SAML single sign-on and SCIM user provisioning (for example Microsoft Entra ID) on Professional and Enterprise.
- Field-level masking of personal information for roles that should not see it.
- Files are only ever served through download links that expire after 15 minutes.
Evidence integrity
- Every uploaded file is fingerprinted with a SHA-256 hash so you can show it has not changed.
- Replacing a document keeps the original as a prior version. Nothing is overwritten.
- An audit log records who created, changed, deleted or exported records, and every AI run.
- Running sheet entries confirmed from Claude connector drafts are locked. Corrections are added as new, linked entries.
Encryption
- All traffic is encrypted in transit with TLS, and the site enforces HTTPS (HSTS preload).
- Data is encrypted at rest with AES-256 by our infrastructure providers.
Backups and recovery
- Point-in-time recovery of the database to any minute in the last 7 days.
- Daily database backups, each kept for 98 days.
- Deleted files remain recoverable for 7 days, and deleted records sit in a recycle bin for 30 days.
- All backups are stored in Sydney.
Retention, export and deletion
- Retention policies by case type, with notice periods and legal holds, to anonymise or delete closed matters on schedule.
- Full workspace export at any time: every record (JSON and CSV) and every file in one archive, with a SHA-256 manifest.
- When a workspace is deleted, admins receive a deletion certificate listing what was deleted and the date the last backup copy expires.
The Privacy Act and APP 8
Under Australian Privacy Principle 8, an organisation that discloses personal information to an overseas recipient generally remains accountable for how that recipient handles it. Because Case360 keeps case data and AI processing in Australia, using Case360 for investigations does not by itself involve sending that information overseas. Our privacy policy explains how we handle personal information.
Frequently asked questions
Is Case360 data stored in Australia?
Yes. Case records, files, sign-in accounts and backups are stored in Sydney on Google Cloud and Supabase, and the application runs on Vercel’s Sydney region.
Does Case360’s AI send investigation data overseas?
No. AI text analysis uses Claude through Amazon Bedrock’s Australian inference profile, and audio, video and transcription use Google Vertex AI or Case360’s own Whisper service in Sydney.
Is our investigation data used to train AI models?
No. Under the Amazon Bedrock and Google Vertex AI service terms that Case360 uses, customer prompts and outputs are not used to train their models, and Case360 does not train models on customer data.
Does any data leave Australia?
Case data does not, unless an admin turns on an optional feature that says so (email-to-case or the Claude connector), both off by default. Technical error data goes to Sentry in Germany, and billing details are held by Stripe.
Can we export all of our data?
Yes. Admins can export the whole workspace, every record and every file, as a single archive at any time, including after cancelling.
What happens to our data when we leave?
The workspace is deleted, admins receive a deletion certificate, and remaining backup copies expire automatically within 98 days for the database and 7 days for files.
Does Case360 support single sign-on?
Yes. SAML single sign-on and SCIM provisioning, including Microsoft Entra ID, are available on the Professional and Enterprise plans.
Need this for a security review?
Email support@case360.com.au and we will help with your vendor questionnaire.