Verify My Stack exists to answer one question with laboratory evidence: does this product contain what its label claims, and nothing it shouldn’t? Members pool dues to fund independent testing of dietary supplements and longevity compounds. This Charter defines how every test is conducted, so that any member, any brand, and any court can verify that our results are accurate, our methods are sound, and our conclusions are limited to exactly what the data supports.
We accept no samples from manufacturers, distributors, or vendors. We accept no payment, sponsorship, advertising, free product, or other consideration from any brand whose products we test or may test. We carry no affiliate links to any tested product. No brand can pay to be tested, pay to avoid being tested, or pay to influence how results are presented. Our only revenue is member dues. If this ever changes in any respect, the change will be disclosed on every affected result page before publication.
Products are selected by member nomination and ranked-choice member vote, conducted monthly. The founder may add selections to the queue for editorial reasons (for example, a newsworthy category), and any such selection is labeled “editor’s pick” rather than “member vote” on the result page.
Every tested unit is purchased at arm’s length, at retail, from the same channels an ordinary consumer would use (manufacturer’s own store, Amazon, or brick-and-mortar retail). We purchase a minimum of two sealed retail units per test. We photograph and log: the order confirmation, the received package, the intact factory seals, the label (all panels), and the batch/lot number and expiration date. We never accept product from any party with an interest in the outcome, including members, for published testing.
From receipt to laboratory submission, custody is documented at every transfer. Units are stored sealed, in manufacturer-recommended conditions, and opened only for sample splitting as described below. A custody log accompanies every sample: who handled it, when, what was done, with photographs at each step. This log is retained for a minimum of three years and made available in any dispute.
Unit A is opened and split under documented conditions: Portion A is submitted to the laboratory; Portion B is resealed, labeled, and retained. Unit B remains factory-sealed and retained. Retained material is stored for a minimum of twelve months after publication and is available for confirmation retesting as described in Sections 8 and 10.
We use only third-party laboratories that meet all of the following: ISO/IEC 17025 accreditation with the relevant analyses in scope; documented participation in inter-laboratory proficiency testing programs; validated methods appropriate to the analyte (for example, HPLC or HPLC-MS for potency and identity, ICP-MS for heavy metals); and no financial relationship with any brand whose product they analyze for us. Samples are submitted blinded — identified by code, not brand name — wherever the analysis permits. We maintain relationships with at least two qualified laboratories at all times, so that no single laboratory is a point of failure or a source of unresolvable dispute.
Before we publish any failing result — defined as a measured active-ingredient content more than 20% below label claim, any heavy-metal result above applicable regulatory or Proposition 65 thresholds, any positive identification of an undeclared active ingredient, or any microbiological failure — the retained Portion B is submitted to a second, independent qualified laboratory for confirmation. We publish a failing result only when both laboratories agree. If they disagree, we test sealed Unit B at a third laboratory and publish all three data sets with a narrative explanation. Passing results are published on a single analysis, with the retained material available for challenge under Section 10.
Every result page includes: the full Certificate of Analysis as received from the laboratory, including chromatograms and raw values; the batch/lot number, expiration date, purchase date, and purchase channel; the complete methodology trail (acquisition photos available on request); and plain-language interpretation that is strictly limited to the tested batch. We do not extrapolate. Our published language takes the form “the tested lot [number] contained X” — never “this brand’s products contain X.” Passing results are published with the same prominence as failing results; the database’s value is completeness, not headlines. We do not publish opinions about companies, their owners, or their intentions — only measurements and the label they were measured against.
Every tested brand is notified at publication and has a standing, permanent right to: (a) submit a written response of any length, which we append to the result page unedited; and (b) fund a confirmation retest of retained material (Portion B or sealed Unit B) at a mutually agreed ISO/IEC 17025 laboratory, with both parties entitled to observe sample transfer. If a brand-funded confirmation contradicts our published result, we publish the new data with equal prominence and revise the result page’s summary accordingly.
If we discover an error — ours or a laboratory’s — we correct the result page prominently and permanently, with a dated note describing what changed and why. We do not delete results. We do not quietly edit. Result pages are never removed in response to legal threats; they are corrected in response to evidence.
The founder and any staff or contractors must disclose any financial interest in any supplement brand, laboratory, or retailer before it becomes relevant to any test, and must recuse from selection and interpretation decisions where a conflict exists. A current conflicts statement is maintained on this site. As of Version 1.0: none.
Verify My Stack does not sell, broker, resell, ship, or take custody of any consumable product on behalf of members or anyone else. We do not facilitate group purchases of products. We do not test or discuss the sourcing of prescription drugs or controlled substances. We do not provide medical advice, and nothing we publish is a recommendation to take, or not take, any substance.
This Charter may be amended only by publishing a new version with a dated changelog entry describing every change. Results published under a prior version remain governed by, and labeled with, the version in force at the time of testing.
Changelog: v1.0 — initial publication, August 2026.