"Third-Party Tested" on a UK Site: What the Phrase Commits the Vendor To
Under the UK advertising code, a testing claim is an objective claim, and the seller must hold evidence for it before publishing. That gives a buyer a fair question to ask, and a way to read the answer.
"Third-party tested" commits a UK seller to one thing enforceably: holding documentary evidence for the claim before publishing it. Under the CAP Code, marketers must hold that evidence for claims consumers are likely to regard as objective and that can be objectively substantiated, and must not materially mislead, including by omitting material information [1]. What the phrase does not commit a seller to is any particular kind of testing. It can cover three quite different arrangements, and the report header usually tells you which.
The general question of what "lab tested" means as a phrase is answered in what "lab tested" actually means as a phrase. This page is narrower and UK-specific: what the UK advertising rules require behind the claim, how to identify the arrangement from the document, and what to ask. It is not a guide to reading results.
What the UK advertising code requires
The CAP Code applies to marketing communications, which includes claims on a seller's own website. Four rules in its section on misleading advertising bear on testing claims [1].
| Rule | What it says, in short | What it means for "third-party tested" |
|---|---|---|
| 3.1 | Marketing must not materially mislead or be likely to | The claim must match what was actually done |
| 3.3 | Marketing must not omit material information | Leaving out that a test covered a different batch may itself mislead |
| 3.7 | Marketers must hold documentary evidence for objective claims before publishing | The report must exist before the claim appears |
| 3.11 | Marketing must not exaggerate capability or performance | An identity test cannot be presented as proof of purity |
The ASA's substantiation guidance adds the practical point: objective claims need evidence that is held before publication and can be produced without delay if challenged, and the evidence must support the claim as a consumer would understand it [2]. "Third-party tested" is plainly objective. It is capable of being true or false. That places it squarely in the category the rule was written for.
What the code does not do is define the phrase more tightly than an ordinary reader would. A seller who had an outside laboratory test a sample of some batch at some point may be able to substantiate "third-party tested" literally. Whether that is misleading depends on what readers take from it in context. That judgement belongs to the ASA, not to the buyer, but it explains why a buyer should ask which arrangement lies behind the words.
The three things the phrase can mean
| Arrangement | Who chose the sample | Who is the laboratory's client | What it shows the buyer |
|---|---|---|---|
| Manufacturer release testing at an outside lab | The manufacturer | The manufacturer | How the batch left production |
| Seller-commissioned test | The seller | The seller | What one vial the seller sent measured |
| Independent testing of the batch sold | Someone other than the seller, or by a documented sampling plan | Varies; often a buyer or a verification service | The closest thing to independent batch evidence |
None of these is dishonest. Manufacturer release testing is the ordinary way chemical batches are released, and a seller commissioning its own tests is doing more than many. The difference is how much weight each can bear for a buyer who was not there. Only the third removes the seller from the choice of sample, and even then the laboratory still reports only on what it received [3].
There is also a fourth use of the phrase, the in-house result presented as third-party. A certificate on the seller's own letterhead, or on the manufacturer's with no outside laboratory named, is in-house testing. That can be good work. It is not third-party testing, and describing it that way is the kind of mismatch rule 3.1 exists to catch.
Telling them apart from the report header
The results section of a report looks much the same in all three cases. The header does not. ISO/IEC 17025 expects a laboratory's report to identify the laboratory, the client, the item tested and its receipt date, and to mark information supplied by the client as such [3]. Read those fields before any number.
- Issuing laboratory: a named laboratory with an address, separate from the seller and the manufacturer. If the only name is the seller's, it is in-house.
- Client or customer: the party that commissioned the test. The seller's name here means seller-commissioned. A buyer's or verification service's name suggests the third arrangement.
- Sample description: often marked as supplied by the client. It records what the client said the sample was, not what the laboratory confirmed.
- Batch or lot reference: it should match the vials on sale. A report with no batch reference cannot be tied to anything you can buy.
- Dates: sample received and report issued. A report older than the batch on sale describes a different batch.
- Accreditation mark and number, if any: a separate claim, checked against the laboratory's schedule on the UKAS register [4].
If the seller shows only a cropped image of the results with the header removed, the three meanings cannot be separated at all. That is worth asking about, politely and directly. Interpreting the results themselves is a different job, covered in the handbook's guide to reading a certificate.
Questions to ask a UK seller
- Which laboratory tested this batch, and can I see the full report, including the header?
- Who was the laboratory's client, and who chose the vial that was tested?
- Does the report's batch reference match the batch I will receive?
- When was the sample received by the laboratory?
- Which tests were run: identity, purity, content, anything else?
- If the report carries an accreditation mark, which lines are within the laboratory's accredited scope?
Because the claim must already be substantiated, a seller making it should have the answers to hand [2]. A quick, complete reply is a good sign. A reply that answers different questions from the ones asked is information too.
Related words that carry the same obligation
"Independently verified", "lab verified", "tested by a UK lab" and "UKAS-tested" are all objective claims in the same sense, and the same substantiation rule applies [1]. Each invites the same follow-up: verified by whom, on what sample, against what. "UKAS-tested" deserves particular care, because UKAS accredits laboratories and does not test products; the phrase can only sensibly mean a test run under an accredited method, and that is checkable on the schedule [4].
The same logic applies in reverse. A seller who publishes the full report, header included, with a batch reference you can match and a laboratory you can look up, has made a claim that could be shown to be false. That is what makes it worth something. Everything short of that asks you to take the words on trust, which is exactly what a substantiation rule is meant to make unnecessary.
If a claim looks unsupported
A buyer's practical remedy is to buy elsewhere or commission their own test. The ASA handles complaints about advertising claims, including claims on company websites, and applies the substantiation rule when it does [2]. Whether a particular claim breaches the code is for the ASA to decide on the facts, and a buyer does not need to reach a view on that to decide whether a report is good enough to rely on.
References
- The CAP Code, Section 03: Misleading advertisingCommittee of Advertising Practice / Advertising Standards Authority
- SubstantiationAdvertising Standards Authority / Committee of Advertising Practice
- ISO/IEC 17025:2017 — General requirements for the competence of testing and calibration laboratoriesInternational Organization for Standardization / International Electrotechnical Commission, 2017
- Search UKAS accredited organisationsUnited Kingdom Accreditation Service (UKAS)