Whether a published document exists, whether a link resolves to a laboratory's own portal, whether a stated price appears at checkout, whether a batch number matches a certificate.
Whether a published document exists, whether a link resolves to a laboratory's own portal, whether a stated price appears at checkout, whether a batch number matches a certificate. These are verifiable by a reader without trusting the seller at all.
Consistency, whether figures published a year ago still match, whether failures were left visible. A reader can check these given a record, which is why a retained archive is worth more than any single page.
Intent, care in handling, what happens to material between testing and dispatch. These are real and consequential, and no document a seller publishes can establish them, which is worth stating plainly rather than papering over.
A result resolving on a laboratory's own domain is harder to fabricate than a file hosted by the seller. The shift is not absolute and it is the single largest improvement available in this category.
A record containing results that reflect badly is costly to produce and nearly impossible to fake convincingly, because the incentive runs the other way. It is the closest thing to a credible signal a seller can send.
An accreditation says a process was examined against a standard. It does not follow a specific batch, and conflating the two overstates what a credential covers.
Effort spent checking unverifiable claims is wasted, and effort spent checking verifiable ones is cheap. Sorting claims into the three categories first makes the checking tractable.
States which category each of its claims falls into, and does not present the third as the first. That is itself a checkable behaviour.
A document hosted on the laboratory's own domain, reachable without going through the seller, removes the seller from the chain at the point where tampering would be easiest. It is not absolute proof and it is the largest single improvement available in this category.
Comparing what a seller published months ago with what it publishes now costs a reader nothing and cannot be faked retrospectively without leaving traces. A seller whose figures have moved without explanation has told you something no current page would.
The claims on this page about the claims that are checkable, and the ones that are structurally unverifiable can be checked against sources outside this site: