Buying Supplements in the UK: How to Spot a Legitimate Seller

Start with the label, then doubt it

Eighty-nine per cent of 57 sports supplements bought online did not contain the ingredients their labels declared. Forty per cent held no detectable trace of the headline ingredient at all. Seven contained compounds prohibited by the US regulator.

That was Pieter Cohen's team in JAMA Network Open, 2023. One category, one country, one purchase per brand. It is not the whole market.

But the pattern repeats whenever somebody runs the assay. The same group tested 25 melatonin gummies for JAMA that year. Twenty-two were mislabelled, and measured content ran from 74% to 347% of the number on the tub.

So treat a label as a claim rather than a measurement. A legitimate seller is one who can show you the measurement.

What UK law already asks of a seller

Britain regulates supplements as food. Three instruments do most of the work, and each leaves a visible trace on a compliant product.

The Food Supplements (England) Regulations 2003 fix the label. The words "food supplement" must appear. So must a stated portion, a warning not to exceed it, a line saying the product does not replace a varied diet, and an instruction to keep it away from young children. A UK business address is required as well.

GB retained Regulation (EC) 1924/2006 governs claims. Only wording authorised on the Great Britain nutrition and health claims register may be used, and only for the named nutrient at the qualifying dose. A paraphrase is not authorised. Neither is an implication.

Assimilated Regulation (EU) 2015/2283 covers novel foods, meaning anything not eaten here to a significant degree before 15 May 1997. That is why NMN's status in the UK remains a live question, and why how the novel foods process runs is worth ten minutes before you buy anything unusual.

A seller who has read all three writes duller copy than their competitors. That dullness is the signal.

The claims that give a bad seller away

On 24 July 2024 the ASA published six rulings on food supplement advertising. All six were upheld. None came from a public complaint — the regulator's Active Ad Monitoring system found them.

The wording that failed is instructive. "Reduced stress and anxiety." "Crush stress." "Prevent future diseases." CAP Code rule 15.6.2 bars any food ad that states or implies action on a named illness, and the ASA reads that rule broadly.

Sleep claims failed for a different reason. There is no authorised claim on the GB register relating to sleep quality, so "deeper sleep" had nothing to stand on. One advertiser lost a claim it was otherwise entitled to make: "reduction in tiredness and fatigue" is authorised for niacin only in the form "contributes to the reduction of tiredness and fatigue". Dropping two words was enough.

The ASA also ruled that "adaptogen", "nootropic" and "antioxidant" are specific health claims, not neutral descriptors. And it flagged turkey tail mushroom as an unauthorised novel food, which makes the product unlawful to sell, never mind advertise.

Separately, the MHRA can determine that a food supplement is functioning as a medicine. Its Guidance Note 8, updated on 12 May 2026, sets out how. A product that crosses that line needs a marketing authorisation it almost certainly does not hold.

Read a seller's copy against those four points. Most failures are visible in under a minute.

Where you buy shifts the odds

The FSA published its 2025-26 retail surveillance results this July. Samplers bought 845 products across England, Wales and Northern Ireland, and 800 were analysed against the relevant law. Overall, 72% were satisfactory.

The breakdown is more useful than the headline. Products from large retailers scored 82%. Products from small food businesses scored 64%. Online purchases scored 65% against 73% for goods bought in person.

One caveat matters. This was targeted sampling of higher-risk lines, not a random survey of the UK food market, so the absolute numbers run low by design. The comparison between channels still holds, because every channel was sampled the same way.

The survey also caught a problem specific to buying online. On 14 occasions the delivered product carried allergen labelling that had not been visible in the listing. What arrives is not always what the page described.

Ask for the certificate of analysis

A certificate of analysis is the lab report for one manufactured batch. It is the only document that connects the number on the tub to a measurement.

A useful one names five things: the batch code printed on your product, the testing laboratory, the assay date, the method, and the measured content of the active ingredient. Heavy metals and microbiology usually appear too. For an oil, ask for peroxide value, anisidine value and the resulting TOTOX figure.

A generic PDF with no batch number is a brochure. So is a certificate dated three years ago. So is a "certificate" issued by the manufacturer's own in-house lab with no accreditation named — look for ISO 17025.

Batch certification programmes are the industrial version of the same idea. Informed Sport requires every batch to pass a prohibited substances screen before release, and LGC runs over 25,000 samples a year against more than 285 compounds. That tells you what is absent. It says nothing about efficacy.

If a seller cannot produce a batch certificate within one working day, they probably do not hold one.

What testing has actually found in the tub

Contamination is not hypothetical. Between October 2000 and November 2001, researchers at the German Sport University bought 634 non-hormonal supplements from 215 suppliers across 13 countries. Ninety-four of them, 14.8%, contained anabolic-androgenic steroids that were not on the label.

That study is old, and manufacturing has improved. It also remains the largest systematic survey of its kind, and it is why batch certification exists at all.

Degradation is the quieter problem. Albert and colleagues tested 32 fish oil supplements in New Zealand. Only three met or exceeded their labelled EPA and DHA content, 69% delivered under two-thirds of it, and half exceeded recommended TOTOX limits. Oxidation is invisible, and what testing finds in fish oil is rarely on the label.

Undeclared additives turn up too. In the FSA survey, two of 40 green powdered supplements contained amaranth, a colour not permitted in supplements, and one contained sorbic acid. None of the three declared it.

The ten-minute check before you pay

Six things, in order of how much they tell you per minute spent.

Read their claims first. Any wording implying action on a named illness ends the assessment. So does a promise about your specific results.

Find the mandatory label text. No "food supplement" wording, no portion warning, no UK address: walk away.

Check one health claim against the GB register. Pick the boldest one on the page and search the register on gov.uk. If it is absent, so is their compliance function.

Ask for the batch certificate. Match the batch code to your tub. Check the lab and the date.

Check the dose against a trial, not against a competitor. A tub can be perfectly compliant and still hold a dose no study has ever used. Knowing what a p-value does and does not tell you helps here more than any review site.

Look for a named responsible person. A UK company number, a real address, a route to a human. Drop-shippers rarely have all three.

Where our own category comes off badly

In 2024, Andrea Maier's group at NUS published an analysis of 23 longevity supplements in GeroScience. Eighteen were NMN. Measured content ranged from undetectable to 99.2% of sample weight, and deviation from label claim ran from -100% to +11.2%.

Minus one hundred per cent means an NMN product with no NMN in it. That is our category, tested by a research group with no commercial interest in the result.

The honest conclusion is narrow. Batch testing tells you the powder is what the label says. It does not tell you the powder does anything, and for NMN the human evidence remains thin and mixed. Both statements are true at once.

That is the standard we would ask of anyone selling to you, including us. Our Omega-3 DHA and NMN + Resveratrol both carry batch codes that match an independent assay. Ask for it. Ask everyone else for theirs.

References

1. Cohen, P. A., Avula, B., Wang, Y. H., Katragunta, K., & Khan, I. (2023). "Presence and Quantity of Botanical Ingredients With Purported Performance-Enhancing Properties in Sports Supplements." JAMA Network Open, 6(8), e2323879. doi:10.1001/jamanetworkopen.2023.23879
2. Cohen, P. A., Avula, B., Wang, Y. H., Katragunta, K., & Khan, I. (2023). "Quantity of Melatonin and CBD in Melatonin Gummies Sold in the US." JAMA, 329(16), 1401-1402. PMID: 37097362. doi:10.1001/jama.2023.2296
3. Geyer, H., Parr, M. K., Mareck, U., Reinhart, U., Schrader, Y., & Schanzer, W. (2004). "Analysis of non-hormonal nutritional supplements for anabolic-androgenic steroids - results of an international study." International Journal of Sports Medicine, 25(2), 124-129. PMID: 14986195. doi:10.1055/s-2004-819955
4. Albert, B. B., Derraik, J. G. B., Cameron-Smith, D., Hofman, P. L., Tumanov, S., Villas-Boas, S. G., Garg, M. L., & Cutfield, W. S. (2015). "Fish oil supplements in New Zealand are highly oxidised and do not meet label content of n-3 PUFA." Scientific Reports, 5, 7928. doi:10.1038/srep07928
5. Tan, C. L., et al. and Maier, A. B. (2024). "Testing the amount of nicotinamide mononucleotide and urolithin A as compared to the label claim." GeroScience, 46. doi:10.1007/s11357-024-01257-2
6. Food Standards Agency (2026). "Retail Surveillance Survey Sampling Programme 2025/26." FSA Research and Evidence. 845 samples purchased across England, Wales and Northern Ireland, 800 analysed.
7. Medicines and Healthcare products Regulatory Agency (2026). "Guidance Note 8: A Guide to What is a Medicinal Product." MHRA, version published 12 May 2026.
8. Committee of Advertising Practice (2024). "CAP Code Section 15: Food, food supplements and associated health or nutrition claims." Advertising Standards Authority. Six food supplement rulings published 24 July 2024, all upheld.
9. Department of Health and Social Care (2025). "Dietary Supplements: Regulation." Written answer UIN 73506, 5 September 2025, UK Parliament.

A label you can check against a lab report

Every batch of our NMN + Resveratrol is assayed by an independent laboratory, and the batch code on your tub matches the certificate. We publish what the analysis found, including the parts that are unremarkable.

Shop NMN + Resveratrol →