Skip to content

Methodology: How We Define and Grade a Receipt

How The Crypto Online defines a receipt, grades a claim against it, and why the desk does not call itself an accredited fact-checker.

The three-row panel

Every verification on this site runs the same fixed panel, in the same order, because consistency is what makes the method checkable:

  • The claim — quoted directly, attributed to whoever made it, and dated.
  • The receipt — the primary record that supports or contradicts the claim.
  • What it doesn’t prove — the inference the receipt cannot support, stated plainly so readers do not read more into a record than it contains.

What counts as a receipt

A receipt has to be a real artefact that another person can go and look at. We treat five kinds as valid:

  • A blockchain transaction hash or contract address, checkable on a public block explorer.
  • A regulatory or corporate filing (for example, a disclosure filed with a securities regulator).
  • A signed attestation, such as a proof-of-reserves report with a verifiable signature.
  • An archived copy of a web page, with the archive’s own capture timestamp, used when the live page has since changed or disappeared.
  • An on-the-record statement from the party making the claim, where no document exists but a direct, attributable quote does.

If none of these exist for a claim, the finding is “no receipt available.” That is not a placeholder — it is the actual result of the check, and we publish it as such rather than leaving the claim unexamined.

How we grade the outcome

Each verification resolves to one of three outcomes:

  • Supported — the receipt is consistent with the claim as stated.
  • Contradicted — the receipt conflicts with the claim, in whole or in part.
  • No receipt — we could not locate a qualifying artefact, and we say so.

These grades describe the relationship between a statement and a record. They are not a trading signal: “supported” does not mean we endorse an asset, and “contradicted” does not mean sell it. Readers who want to act on anything they read here should treat it as one input among many, verified again through their own judgment — see our disclaimer.

A 403 is not a 404

When a source exists but blocks automated or even manual access, we record it as blocked, not as absent. The distinction matters: a blocked source may still be real and checkable by a human with the right access, while a genuinely missing source means the claim simply has nothing behind it. We try to say which situation we are in rather than collapsing both into “we couldn’t find it.”

Pulling the receipt yourself

Every verification links to the record it relies on so a reader can check our work directly, not just take our word for the grade. Our how-to-verify guides walk through the mechanics: reading a transaction on a public explorer, searching a regulator’s filing database by entity name, finding a page’s archived history, and reading a proof-of-reserves attestation for what it actually attests to.

Why the grade can change

A receipt that looked solid at publication can later be superseded — a filing gets amended, an attestation gets withdrawn, an archived page turns out to have been altered before the crawl. When that happens to a published verification, we do not quietly edit the old conclusion. We log the change, with a date and an explanation, in our corrections log, so the record of what we said and when stays intact.

What this is not

We are not an accredited fact-checking organisation and we do not apply ClaimReview-style ratings borrowed from that world. We are a small editorial desk applying one disclosed method consistently, and we would rather be clear about that boundary than imply a credential we do not hold.

Not financial advice. Verification outcomes are judgments about evidence, not recommendations 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.