Scope
These guidelines apply to every verification, how-to-verify guide, glossary entry, article and newsletter issue we publish. They sit alongside our ethics policy, which focuses on independence and conflicts of interest, and our methodology, which sets out exactly how a receipt qualifies and how outcomes are graded.
Core principles
Three rules apply to everything we publish. Every claim we repeat is quoted, attributed to whoever made it, and dated. Every receipt we cite is a real, linkable artefact — never a paraphrase of one, never a description of a document we have not actually looked at. And every piece is read and signed off by a human editor before it goes live, regardless of how it was drafted.
Sourcing standards
We check claims against primary records wherever one exists: on-chain data, regulatory filings, signed attestations, or archived snapshots. Secondary reporting can point us toward a claim worth checking, but it is never the receipt itself — we trace back to the primary source before we publish a grade.
We do not quote unnamed “sources familiar with the matter” to support a factual claim. If a source needs to stay anonymous, we say so and explain why, and we weigh that against whether the piece can stand without it.
What we do not cover as editorial
We do not run presale or token-launch coverage, price targets, or “best coin to buy” style content as editorial material. Any commercial relationship — sponsorship, paid placement, affiliate arrangement — is disclosed and kept visibly separate from verification and reporting content; it never influences which claims we choose to check or how a receipt is graded.
Disclosure of AI assistance
Where a piece has used AI tools in drafting, research or formatting, we disclose that. AI assistance never substitutes for checking a primary source, and it never generates a receipt, a quote, or a grade on its own — a human editor verifies the underlying record before anything is published.
Balance in the verification device
Our three-outcome grading (supported, contradicted, no receipt) describes the relationship between a claim and a record, not a view on whether an asset is worth holding. We apply the same scrutiny regardless of whether a claim is bullish or bearish for the asset involved, and we do not let a “supported” grade read as encouragement to buy or a “contradicted” grade read as encouragement to sell.
Review before publication
A piece does not go live until an editor other than the contributor has checked that every claim is attributed and dated, every receipt link actually resolves to the artefact it is cited for, and the “what it doesn’t prove” line honestly limits the finding rather than overselling it.
Corrections and accountability
When we get something wrong, we fix it in the open. Our corrections log is dated and append-only: we do not silently edit a published conclusion. Readers can flag an error through our contact page at any time, and correction reports are prioritised over most other mail.
Independence
Editorial decisions — what we check, how we grade it, and whether we run a correction — are made by the editorial desk alone, independent of advertising or partnership arrangements.
Not financial advice. Material published under these guidelines is verification and education, not a recommendation to buy, hold or sell any asset.
Not financial advice. Human-edited; AI-assisted drafting is disclosed in our editorial guidelines. Spotted an error? See how corrections work.