Admiralty Code (Source and Information Grading)
Type: Standardized intelligence grading system
Origin: UK Royal Navy/Admiralty, adopted by NATO; widely used in Western intelligence communities
NATO reference: STANAG 2511; also referenced in PHIA standards (UK professional head of intelligence analysis)
Grading Structure
The system grades source reliability and information accuracy independently using a letter+number matrix:
Source Reliability (Letter A–F)
| Grade | Meaning |
|---|---|
| A | Completely reliable — consistent, verified track record |
| B | Usually reliable — minor anomalies or short track record |
| C | Fairly reliable — more often right than wrong |
| D | Not usually reliable — more often wrong than right |
| E | Unreliable — consistently wrong or manipulative |
| F | Cannot be judged — insufficient history |
Information Accuracy (Number 1–6)
| Grade | Meaning |
|---|---|
| 1 | Confirmed by other sources — independently corroborated |
| 2 | Probably true — consistent with established facts |
| 3 | Possibly true — not corroborated |
| 4 | Doubtful — inconsistent with established facts |
| 5 | Improbable — contradicted by established facts |
| 6 | Cannot be judged — insufficient information |
Application for Open-Source Analysts
In the OSINT context, the code adapts:
- Source grades apply to publication reputation, editorial standards, track record, funding transparency
- Information grades apply to specific claims: corroborated via independent primary source = 1; single source only = 3; contradicts documented facts = 5
- State media (RT, Xinhua, IRNA): typically D or E on source grade — not “unreliable” as a factual matter but systematically optimized for political messaging
- Social media / anonymous Telegram channels: typically F + 3 (cannot judge reliability; claim needs corroboration)
Key Connections
- Source Verification Framework
- Intelligence Confidence Levels
- 04 — Source Evaluation Without Institutional Context
Related
- Intelligence Confidence Levels — source/information grades feed the calibrated-confidence layer of a finished judgment
- Analysis of Competing Hypotheses — graded evidence enters the ACH matrix; reliability gates diagnosticity
- Cognitive Biases in Intelligence Analysis — explicit grading counters the volume and vividness biases that inflate weak sources