A source is not a finding. Before any statement enters a Greyline report, it is graded for reliability and credibility using a modified NATO Admiralty Code. Here is why that discipline matters and what it looks like in practice.
Methodology noteAnalysis
Every investigation produces material of wildly varying quality. A signed court judgment and an anonymous forum post might both mention the same name, but they are not the same kind of evidence, and a report that treats them identically has already failed the client reading it.
This is the problem source reliability grading exists to solve. It is not a formality. It is the mechanism that stops a weak signal from being dressed up as a strong one.
Greyline grades sources using a modified version of the NATO Admiralty Code, a two-axis system originally developed for military intelligence assessment and now widely used across the private and government intelligence sectors.
Every source is scored on:
The two axes are scored separately because they answer different questions. A generally reliable source can still report something dubious. A source with no track record can still produce something independently verifiable. Collapsing the two into a single “trust score” loses exactly the distinction that matters.
A source graded A1 (completely reliable, confirmed) supports a stated fact. A source graded E5 (unreliable, improbable) does not belong in a finding at all, no matter how convenient it would be to include it.
Grading is not an internal exercise that stays in an analyst’s notebook. Where it is material to a conclusion, the grade is disclosed in the report itself.
In practice, this means a Greyline report distinguishes between three different things that a less disciplined report would present as equivalent:
A client, a solicitor, or a court reading a Greyline report can see not just what was found, but how much weight the finding can bear. That distinction is the entire point of grading, and it is why our full breakdown of finding classification lives on the Analysis page of our published method.
Grading disciplines are easy to state and easy to abandon under pressure. A tight deadline, a client who wants a stronger answer than the evidence supports, or a source that is inconvenient to downgrade all create pressure to round up.
The discipline only holds if it is applied the same way regardless of what the client wants to hear. A clean finding is reported as a clean finding. An unresolved question is reported as unresolved. Grading a weak source generously because the conclusion would otherwise be less satisfying is not analysis, it is storytelling with a badge of rigour attached.
This is also why grading has to be recorded contemporaneously, in the working notes, at the point the source is used, not reconstructed afterwards to justify a conclusion already reached.
If you are instructing an investigator, whether Greyline or anyone else, the questions worth asking are simple: does this firm grade its sources at all, is that grading visible in the final report, and does a single-source claim look different on the page from a corroborated fact?
If the answer to any of those is no, the report may still be accurate. But you have no way to tell the difference between a well-supported conclusion and a confident guess, and that is precisely the gap source reliability grading is designed to close.