Almost every supplement brand now says its products are "lab tested". Very few show you the test. A certificate of analysis is the document behind that phrase, and once you know which eight lines to look at, it takes about a minute to tell a real one from a decorative one.
The short answer
A certificate of analysis, usually shortened to CoA, is a laboratory report for one specific batch of one specific product. It is only meaningful if it names the batch, comes from a named laboratory, states the method used, and shows a measured result next to the specification it was tested against. A certificate without a batch number is not evidence about the tub in your hand. A "pass" is also not a claim of purity: it means the result sat under a legal or internal maximum, not that nothing was found. The phrase "lab tested" on a label, on its own, is unverifiable, which is exactly why it is so popular.
What a certificate of analysis actually is
A CoA is a snapshot. A sample was taken from one production batch, sent to a laboratory, tested for a defined list of things, and the results were written down. That is the whole document. It does not cover the batch made last month, it does not cover a different flavour or size, and it does not certify the brand as a company.
Three families of test usually appear on it:
- Identity and potency. Is the declared substance actually present, and in the amount stated per portion?
- Contaminants. Heavy metals such as lead, cadmium and mercury, sometimes solvent residues, pesticide residues or mycotoxins depending on the raw material.
- Microbiology. Total counts, yeasts and moulds, and specific organisms such as Salmonella or E. coli.
A certificate that only shows one of the three is not wrong, but it is partial, and it should be read as partial.
The eight lines that decide whether a certificate means anything
1. The batch or lot number
This is the line that connects the paper to the product. Find the lot code printed on your tub or pouch and check that it matches the certificate. If the certificate carries no batch number at all, it cannot be linked to anything you own. Batch traceability is not an optional courtesy either: Article 18 of Regulation (EC) No 178/2002 requires food businesses to be able to trace food at all stages of production, processing and distribution.
2. The exact product and form
"Magnesium" is not a specification. "Magnesium bisglycinate powder" is. The certificate should name the material with the same precision the label uses, because the permitted chemical forms are themselves regulated: Directive 2002/46/EC allows only the vitamins and minerals listed in its annexes, in the forms listed there, and purity criteria for those forms are set at EU level. We looked at that distinction in detail in our piece on which supplement label words are legally defined.
3. The date of analysis
A certificate from three years ago tells you about a batch that has probably long since been consumed. The analysis date should sit close to the manufacturing date of the batch you bought.
4. The laboratory, by name
A report with no laboratory name, no address and no signature is a leaflet. An in-house laboratory is not automatically a problem, and many good manufacturers run one, but "tested in our own facility" and "tested by an independent laboratory" are different statements and should not be blurred.
5. Accreditation, and its scope
The relevant standard is ISO/IEC 17025, the international standard for the competence of testing and calibration laboratories. Accreditation to it is granted by a national accreditation body and covers a defined scope: specific methods, specific matrices, specific elements. A laboratory can be accredited for heavy metals in water and not for heavy metals in a capsule blend. So the useful question is not "are you accredited" but "is this test inside your accredited scope".
6. The method
Serious certificates name the technique. Heavy metals are typically measured by ICP-MS, inductively coupled plasma mass spectrometry, which is the method used in the published surveys cited below. Vitamins are often measured by HPLC. If the method column is blank, nobody can judge whether the number is credible.
7. Specification versus result
These are two different columns and they are frequently confused. The specification is the limit the product was tested against. The result is what the instrument measured. A certificate that prints only specifications, with no measured values, is a wish list. If you see a tidy table of limits and no numbers beside them, that is the whole finding.
8. Units, and the limit of detection
Milligrams per kilogram and micrograms per kilogram differ by a factor of a thousand, and mixing them up turns a reassuring number into an alarming one. More subtle: "not detected" is not the same as zero. It means the substance was below the limit of detection of that method on that day. A good certificate prints the limit of detection or limit of quantification next to the result, so "not detected" has a size.
What the law already requires, and what it does not
Testing does not happen in a vacuum. Maximum levels for lead, cadmium and mercury in food supplements have existed in EU law since July 2009, and today the binding EU wide provisions sit in Commission Regulation (EU) 2023/915 on maximum levels for certain contaminants in food, which replaced Regulation (EC) No 1881/2006. The German federal environment and consumer protection ministry and the Food Safety Authority of Ireland both describe that framework in plain terms.
Two consequences follow, and they matter more than the numbers themselves.
First, the maximum levels apply to the ready to use supplement, not to each individual raw material, and the exact figure depends on the product category, so a supplement based mainly on dried seaweed or on minerals is not judged by the same value as a capsule of a purified vitamin. If you want the current figure for a given category, read it in Annex I of the regulation rather than trusting a round number quoted in a blog post, including this one.
Second, a compliant result is a legal ceiling, not a promise of absence. Everything grown in soil carries traces of metals. A certificate showing a measurable but low lead value is a normal document, not a scandal.
What testing actually catches
The reason any of this exists is that the failure modes are real and documented.
Undeclared pharmaceutical ingredients. An analysis of United States Food and Drug Administration warnings from 2007 through 2016 identified 776 adulterated dietary supplements implicating 146 companies. Most were marketed for sexual enhancement (45.5%), weight loss (40.9%) or muscle building (11.9%), and the most common hidden ingredients were sildenafil, sibutramine and synthetic steroids (Tucker and colleagues, 2018). A separate analysis found that banned drugs were still present in supplements bought after those products had been recalled (Cohen and colleagues, 2014).
Amounts that do not match the label. In a 2023 analysis of melatonin gummies sold in the United States, 25 products were tested. Twenty two of the 25, that is 88%, were inaccurately labelled, and only 3 contained a quantity within 10% of the declared amount. Measured melatonin ranged from 74% to 347% of the label. One product contained no detectable melatonin at all, but did contain 31.3 mg of cannabidiol (Cohen and colleagues, 2023).
Note what those two failure types have in common. They cluster in categories built on dramatic promises. That is a useful filter in itself, and it pairs with the questions we set out in our guide on how to spot a weak study behind a supplement claim.
What the heavy metal numbers usually look like
Contamination coverage tends to arrive in two flavours: nothing to see, or panic. The measured literature sits in between, and it is worth reading properly.
A 2023 study measured arsenic, cadmium, mercury and lead by ICP-MS in children's multivitamin and multimineral products on the Jordanian market. Mean concentrations were 32.4 micrograms per kilogram for arsenic, 58.2 for cadmium, 42.2 for mercury and 231 for lead, with lead ranging from 8.6 to 541 micrograms per kilogram. Estimated daily intakes from those products were all below the tolerable intake limits, and the calculated hazard indices were below 1 (Ababneh, 2023).
A 2026 study analysed 30 sports protein supplements on the Chinese market. Every sample contained at least one target metal. Plant derived products carried higher mean lead (0.18 mg/kg) and cadmium (0.05 mg/kg) than dairy derived products (0.06 and 0.01 mg/kg). One brown rice and cocoa flavoured product stood out with 0.60 mg/kg lead. Modelled hazard index was 0.45 at 30 g per day and 0.90 at 90 g per day, with a 95th percentile of 1.28 under heavy use (Chen and colleagues, 2026).
Three honest readings of that. Traces are normal and usually small. The raw material, not the brand story, drives the difference: plant and mineral derived ingredients carry more than dairy derived ones, which is also why purification and testing matter so much for a natural resin, as we explained in our article on what shilajit and fulvic acid actually are. And dose scales exposure: the risk numbers rise with grams per day, which is why a 30 g scoop deserves more scrutiny than a 1 g capsule serving.
Certificates that should lower your confidence
- No batch number. The single most common weakness. It cannot be tied to your product.
- Specifications only. A table of limits with no measured results.
- An image with no laboratory name. A photograph of a stamp is not a report.
- Raw material certificate presented as a finished product certificate. An ingredient supplier's document says nothing about what happened during blending, encapsulation and packing.
- "Meets all applicable standards". Which standards, measured how, at what value?
- Accreditation logos without a scope. A logo is not evidence that this particular test was inside the accredited scope.
- An old date on a current batch. If the analysis predates the production run, it belongs to a different batch.
How to ask for one, and what to expect
Write to the brand with the lot number from your pack and ask for the certificate of analysis for that batch, with the laboratory named and the methods stated. Three replies are common. Some brands send it. Some send a redacted version, usually because the certificate reveals the contract manufacturer, which is a commercial rather than a safety reason. Some do not answer, and that is information too.
One further note on storage, because it is the part a certificate cannot cover. A test result describes the product as it left the factory. What happens in a warm bathroom afterwards is a different question, which we covered separately in does storage ruin your supplements.
How this applies to us
The test is only worth publishing if it survives being pointed at our own labels. Our product pages carry the phrases "Lab-tested in Germany" and "Independently lab-tested". By the standard set out above, those phrases are a promise from us, not a verifiable fact for you, because we do not currently publish per batch certificates on the website.
What we do publish is the part the law makes checkable: the exact form of every nutrient, the amount per daily portion and the percentage of the nutrient reference value, on pages such as our magnesium complex, our zinc picolinate capsules and our purified shilajit resin. If you want the certificate for the batch you received, write to us with the lot number and ask. Applying this article's test to us is the point of writing it.
Frequently asked questions
Is a certificate of analysis legally required for supplements sold in the EU?
A certificate does not have to be given to consumers. What the law requires is that the product complies: only permitted nutrient forms may be used under Directive 2002/46/EC, contaminants must stay within the maximum levels of Regulation (EU) 2023/915, and the batch must be traceable under Article 18 of Regulation (EC) No 178/2002. A certificate is the ordinary way a business demonstrates it has checked those things.
Does "not detected" mean there is none in the product?
No. It means the amount was below the limit of detection of the method used. A good certificate prints that limit next to the result, so you can see how small "none" actually is.
What does ISO/IEC 17025 accreditation prove?
It is the international standard for the competence of testing and calibration laboratories, granted by a national accreditation body for a defined scope of methods and materials. It says the laboratory is competent for the tests inside that scope. It does not say the product is good, and it does not automatically cover every test the laboratory offers.
My certificate shows a small amount of lead. Should I be worried?
A measurable trace is normal in anything derived from soil or minerals, which is why the law sets maximum levels rather than requiring zero. The useful comparison is the measured result against the maximum level for that category, and the amount you actually take per day. If a result sits at or above the legal maximum, ask the brand about it directly.
Is an in-house laboratory result worthless?
No, but it answers a narrower question. In-house testing is normal in good manufacturing practice. Independent testing removes the conflict of interest. The distinction only matters if a brand blurs the two, so read which one the certificate describes.
Which supplements deserve the most scrutiny?
Two groups stand out in the published record: products sold on dramatic promises, where undeclared pharmaceutical ingredients cluster, and products taken in large daily amounts of plant or mineral derived material, where trace contaminants scale with the serving size.
The Bottom Line
"Lab tested" is a claim. A certificate of analysis is evidence, but only when it names the batch, the laboratory, the method, the specification and the measured result, with units and a detection limit you can see. Read those eight lines and most of the marketing dissolves. Ask for the certificate that matches your lot number, compare the result with the limit rather than with zero, and treat a brand's willingness to answer as part of the answer.
Sources
- Commission Regulation (EU) 2023/915 of 25 April 2023 on maximum levels for certain contaminants in food and repealing Regulation (EC) No 1881/2006. European Commission, 2023.
- Lead, Cadmium, Mercury: Heavy Metals. German Federal Ministry for the Environment, Climate Action, Nature Conservation and Nuclear Safety.
- Metals: regulatory information on contaminants in food. Food Safety Authority of Ireland.
- Directive 2002/46/EC on food supplements, Article 4: permitted vitamins and minerals, permitted forms and purity criteria. European Parliament and Council, 2002.
- Regulation (EC) No 178/2002, Article 18: traceability. European Parliament and Council, 2002.
- ISO/IEC 17025, General requirements for the competence of testing and calibration laboratories: joint ISO, ILAC and IAF communiqué. International Laboratory Accreditation Cooperation.
- Tucker J, Fischer T, Upjohn L, Mazzera D, Kumar M. Unapproved pharmaceutical ingredients included in dietary supplements associated with US Food and Drug Administration warnings. JAMA Network Open, 2018.
- Cohen PA, Maller G, DeSouza R, Neal-Kababick J. Presence of banned drugs in dietary supplements following FDA recalls. JAMA, 2014.
- Cohen PA, Avula B, Wang YH, Katragunta K, Khan I. Quantity of melatonin and CBD in melatonin gummies sold in the US. JAMA, 2023.
- Ababneh FA. Toxic elements in children's multivitamin-multimineral supplements and the health risks they pose. Regulatory Toxicology and Pharmacology, 2023.
- Chen M, Dong R, Lai W. Heavy metal and microplastic exposure from sports protein supplements: integrated health risk modeling and scenario analysis. Environmental Research, 2026.
Note on sources 4 and 5: EUR-Lex blocks automated retrieval, so those two EU instruments are linked through the legislation.gov.uk mirror of the text. The EU instruments themselves are named in full.
