Methodology
How Iron Command assesses capability.
One intelligence-tradecraft framework. READ is how Iron Command reads a capability claim and turns it into a decision — applied with the sourcing standards of an operational intelligence shop. A practitioner tool, not an academic model.
The standards
IC tradecraft
The standards READ runs on. Sources graded on the NATO Admiralty scale, confidence rated under ICD-203, and the competing read tested before any call — the way an operational J2 shop builds an estimate. Applied on every published comparison and every advisory deliverable.
The framework
READ
The framework for reading a capability claim. Reality, Evidence, Alternatives, Decision. Built for defence-tech investors, programme leads, and corporate strategy teams who need to think like an intelligence analyst about defence questions without being one.
The standards behind it
How the rigour is enforced
READ is the framework; these are the intelligence-community standards it runs on. They apply to every comparison on this site and every commissioned advisory engagement.
01
Graded sourcing
Every source weighted by reliability and collection type on the NATO Admiralty scale. Manufacturer claim, state claim, OSINT, and analyst inference never carry equal weight.
02
Rated confidence
Judgements carry explicit confidence under ICD-203, and intelligence gaps are stated rather than hidden. No hedging by default, no false certainty.
03
Tested alternatives
The competing read is steelmanned before any call is made — Heuer's analysis of competing hypotheses. The step commercial analysis skips.
Buyer-side method
READ — how to think like an intelligence analyst about defence claims
Defence capability is systematically misunderstood by the people making decisions about it, because intelligence tradecraft is treated as a classified specialism rather than a transferable methodology. READ is the transferable version. Four questions, in order, every time a capability claim crosses your desk.
R
Reality
What was actually built versus what is being sold. The demo-to-deploy gap. Operational reality versus marketing reality.
E
Evidence
Sourcing, weighting, and confidence calibration. Manufacturer claim, state claim, Western analyst inference, OSINT — each carries different weight, and a buyer reading the brief should know which is which.
A
Alternatives
Test the competing read before committing to one. Steelman the rival explanation, surface the assumptions each requires, and state what would have to be true for the obvious answer to be wrong. Most analysis runs from evidence straight to a conclusion — this is the step in between.
D
Decision
Communicate intelligence to decision-makers, not to other analysts. The report exists to support a decision. If it doesn't, it's a paper, not an assessment.
Methodology in public
Where to see READ on real questions
Every published comparison on this site is a worked example of the same methodology applied inside advisory engagements. The free sample isn't a teaser — it's the audition.
Worked examples
Peer-platform comparisons
READ applied to head-to-head capability questions. F-22 vs J-20, Burke vs Type 055, Virginia vs Yasen-M.
Long-form
Programme and sector reads
READ applied to UK/NATO defence-AI procurement, sector reads, and adversary doctrine.
Reference
Intel Wiki
The open-source intelligence database that underpins every comparison. 170+ platforms across 21 nations.
Led by
Ben Brand
Founder, Iron Command · Former British Army intelligence analyst
Apply the method
The methodology you've just read, applied to your decision.
Every Iron Command Advisory engagement runs READ end to end — graded sourcing and tested alternatives behind closed doors, a decision-shaped deliverable in front of you. Discovery calls within 48 hours.
Built on top · Coming September 2026
First Island Brittleness (FIB) — the index the methods produce.
READ is how we run the assessment and how the buyer reads it. FIBis the public quarterly artefact those methods produce — a 0–100 index of how fast the US-allied Indo-Pacific position is decaying versus being rebuilt. Methodology v1.0 locked, weights immutable for twelve months, open-source inputs per ICD 203.
See FIB →