Risk Assessment Frameworks
| Title | Risk Assessment Frameworks |
| Doc# | DOC-RISK-002 |
| Version | 1.1 |
| Date | 05-03-2025 |
This document sets out how Tessera assesses information security risk. The method draws on ISO/IEC 27005 (information security risk management) and NIST SP 800-30 (risk assessment), and is the engine behind the risk register (DOC-SEC-001), the risk treatment plan and the Statement of Applicability (DOC-SEC-003). It supports Clause 6.1 of ISO/IEC 27001:2022.
Method
- Establish context. Define the assessment scope and the assets within it; agree the risk criteria, including Tessera’s risk appetite and tolerance.
- Identify risk. For each asset, identify the threats that could act on it and the vulnerabilities they could exploit, alongside the controls already in place.
- Analyse risk. Combine likelihood and impact to produce an inherent rating, then re-rate against the controls in place to give a residual rating.
- Evaluate risk. Compare residual ratings against the risk criteria; prioritise what exceeds appetite.
- Treat risk. Select a treatment — mitigate, transfer, accept or avoid — and assign an owner and a date.
- Monitor and review. Reassess on a schedule and after material change or incident.
Rating matrix
Likelihood and impact are each rated on a four-point scale. The rating bands drive prioritisation and reporting to the Executive Risk Committee.
| Likelihood ↓ / Impact → | Low | Medium | High | Critical |
|---|---|---|---|---|
| Rare | Low | Low | Medium | Medium |
| Unlikely | Low | Medium | Medium | High |
| Possible | Medium | Medium | High | High |
| Likely | Medium | High | High | Critical |
Working template
| Risk ID | Asset | Threat | Vulnerability | Controls in place | Likelihood | Impact | Residual | Treatment | Owner | Status |
|---|---|---|---|---|---|---|---|---|---|---|
| RR-01 | Tenant data (Aurora) | Unauthorised access via leaked static credential | Long-lived access keys; absent control-plane/data-plane segmentation | Short-lived credentials (in progress); secret scanning (in progress) | Medium | High | Medium | Mitigate — eliminate static keys; complete segmentation | CISO | In progress |
| RR-02 | Tenant data | Cross-tenant exposure | Shared cluster failure of RLS | Row-level security; tenant-scoped roles | Low | Critical | Low | Mitigate (sustained) | Head of Engineering | Ongoing |
| RR-03 | Tenant data | Data loss or corruption | Backup failure | Aurora continuous backups; restore tests | Low | High | Low | Mitigate (sustained) | Cloud Service Ops | Ongoing |
| RR-05 | Platform egress | Delayed detection of exfiltration | Untuned alert routing | CloudTrail, GuardDuty, VPC Flow Logs, SIEM | Medium | High | Medium | Mitigate — retune management-plane alerts; egress baseline | CISO | In progress |
Using the framework
The register is the live artefact; this document is the method behind it. New risks are logged as they are identified — through threat modelling, vendor assessments, audit findings or incidents — and walked through the analyse → evaluate → treat → monitor cycle. TSR-INC-2025-031 is the clearest recent example: a control gap surfaced by an incident became tracked risks (RR-01, RR-05) with owners and treatment dates, rather than a one-off fix.
REVIEWER NOTE: Several “residual” ratings in the register still reflect controls that are in progress rather than closed. A residual rating is only credible once the treating control has been substantiated; treat in-progress mitigations as not yet effective when reading the register.