Security built for crisis-sensitive data
When a Reddit thread is escalating, the last thing your team needs is a data concern slowing down the response. Defusely is built so security is invisible during a crisis and airtight when audited after one.
Encryption everywhere
Your crisis data is encrypted at every layer.
In transit
All data between your browser and Defusely is encrypted via TLS 1.2+. No exceptions.
At rest
Every War Room, draft, approval, comment, and attachment is encrypted using AES-256. Crisis-sensitive data never sits unprotected.
Key management
Encryption keys are managed by our infrastructure provider with automatic rotation. No keys are stored in application code.
Access control built at the database level
Every database query is scoped to your organization. Users can only see War Rooms, brands, and incidents that belong to their org. This is enforced at the database level, not the application level. Even if the application has a bug, the database rejects unauthorized access.
Role-based permissions
| Role | What they can do |
|---|---|
| Editor | Create and edit War Rooms, drafts, and post-mortem notes |
| Approver | Everything Editors can do, plus approve response drafts before posting |
| Viewer | Read-only access to War Rooms and reports. Cannot edit or approve. |
Shared War Room links
When you share a War Room with an external stakeholder, they get a read-only link with an expiration date and optional password. No account required. No data leakage.
Immutable audit trail
Every action in Defusely is logged with a timestamp, user identity, and description:
Immutable and exportable
This trail is immutable. It cannot be edited or deleted by any user, including admins. The audit trail powers the post-mortem report.
Hardened infrastructure
Security by design, not by feature add-on.
Hosting
SOC 2-aligned cloud infrastructure with automated backups, failover, and disaster recovery.
Database
PostgreSQL with row-level security policies enforced on every query. No direct database access exposed to users or the public internet.
Edge functions
Server-side logic runs on isolated edge functions. Each function handles one job. No function has access to another function's data.
Backups
Automated daily backups with point-in-time recovery.
Input validation and bot protection
Server-side validation
All inputs are validated on the server before being stored, preventing injection attacks and malformed data.
reCAPTCHA Enterprise
Protects forms and sensitive endpoints from automated abuse and bot attacks.
URL validation
Reddit URLs are validated and normalized before processing to prevent tampering.
What we store and what we don't
We DO store
- War Room data and incident summaries
- Response drafts and approvals
- AI analysis and severity scores
- Team member names and roles
- Audit trail and activity logs
We NEVER store
- Passwords (hashed by auth provider)
- Credit card numbers
- Raw Reddit API credentials
- Unencrypted backups
- Personal data beyond what's needed to operate
Data retention
30 days after you cancel your subscription. After that, all data is deleted from live systems.
Data portability
Export all your incident history, reports, and documentation as PDF or structured data anytime.
Compliance and standards
| Standard | Status |
|---|---|
| SOC 2 Type II | Infrastructure aligned. Formal audit planned for 2026. |
| GDPR | Data processing compliant. EU data residency available on Scale plan. |
| CCPA | Consumer data request handling in place. |
| Encryption at rest | AES-256 via infrastructure provider. |
| Encryption in transit | TLS 1.2+ on all connections. |
Questions about security?
Our team is here to discuss compliance, architecture, and any security concerns.
Contact us about security