Start here, because it is the important sentence

The National Institutes of Health Office of Dietary Supplements puts it about as plainly as it can be put:

"These seals do not guarantee that a product is safe or effective."

That is a government source describing voluntary quality seals. Hold onto it, because everything below sits inside it. A certification mark is a statement about what is in the bottle and how it was made. It is not a statement that taking the contents will do anything for you.

The second important thing: FDA does not endorse any of them

There is no federal certification program for dietary supplements, and FDA does not require or endorse any private mark.

FDA states that it "does not approve dietary supplements or their labeling" and "does not test dietary supplements before they are sold to consumers." The closest the agency comes to acknowledging third-party testing is a neutral note that consumers "may contact the dietary supplement manufacturer or a commercial laboratory for an analysis of a product's content," naming no program.

So when you see a certification mark, you are looking at a private, voluntary arrangement between a company and a testing organization. That is not a criticism: in a system with no premarket gate, voluntary external testing is meaningfully better than nothing. It is simply not a government stamp, and nobody should read it as one.

USP Verified

Run by the United States Pharmacopeia, a nonprofit that writes quality standards for medicines and supplements.

USP describes its Verified Mark as confirming that the product:

• "Contains the ingredients listed on the label, in the declared potency and amounts"
• "Does not contain harmful levels of specified contaminants"
• "Will break down and release into the body within a specified amount of time"
• "Has been made according to FDA current Good Manufacturing Practices using sanitary and well-controlled procedures"

Two features make this one distinctive. First, it includes a manufacturing facility audit, against USP's own manufacturing practices chapter and against the federal cGMP regulation at 21 CFR Part 111. Second, it includes a disintegration standard: the product has to break apart in a specified time. That addresses a real failure mode. A tablet that passes through you intact delivers nothing regardless of what is in it.

What it does not address: whether the ingredient does anything. USP makes no efficacy claim.

Participation is voluntary and manufacturer-initiated. USP publishes a list of participating products.

NSF: two different things with similar logos

This is where most confusion happens, because NSF runs two programs and they cover very different ground.

NSF/ANSI 173

NSF describes this as "the only American National Standard for testing and certifying dietary supplements." It covers:

• Label-claim verification: that the contents match the panel
• Contaminants: lead, mercury, arsenic, cadmium and hexavalent chromium
• Microbial testing: aerobic organisms, yeast, molds, Enterobacteriaceae, plus Salmonella, E. coli and S. aureus
• Mycotoxins and pesticides for botanical materials

On frequency, NSF says: "We conduct annual audits and periodically retest certified products to ensure continued compliance."

Read that carefully: annual audits and periodic retesting, not every lot. That is a reasonable and normal certification model. It is simply not the same as lot-by-lot testing, and the distinction matters for how much a mark on a given bottle tells you about that bottle.

NSF Certified for Sport

This is a narrower, more intensive program aimed at athletes subject to drug testing. It screens for substances banned by major sporting organizations, and it does so lot by lot. A product must hold the underlying NSF certification first.

It is recognized or recommended by a number of sports bodies, including USADA and several professional leagues.

One note on the number of substances screened: NSF's own pages currently give different figures in different places: 290 in one place, "295+" in another, "more than 300" in a third. The honest version is "more than 290." If a precise number matters to you, ask NSF rather than trusting a number quoted second-hand, including this one.

What Certified for Sport does not tell you is anything about whether the product is effective. Beyond the underlying certification, it is specifically a banned-substance program, not a general quality program.

Informed Sport and Informed Choice

Both are run by LGC, and the difference between them is the cleanest illustration of the lot-testing question in the whole category. In the program's own words:

Informed Sport: "every batch/lot of a product is tested prior to being released to the market." Aimed at "elite athletes and drug-tested personnel, such as the military, subject to strict doping control standards."

Informed Choice: "At least twelve lots/batches of the product are tested at random every year, but unlike Informed Sport, it is not guaranteed that every batch has been tested." Aimed at "those living an active lifestyle."

The same organization and similar-looking marks carry materially different promises. If you compete in a tested sport, that difference is the entire point. If you do not, Informed Choice still tells you the product is in a sampling program, which is more than most products can say.

Both are banned-substance programs. Neither is a broad quality certification, and neither addresses efficacy.

Banned Substances Control Group

BSCG screens for a larger list: over 450 substances, combining the WADA prohibited list with around fifty prescription, over-the-counter and illicit drugs that are not banned in sport.

The structure is worth understanding, because it is layered differently from the others. BSCG describes testing every finished product lot for banned substances, but annual verification of label claims and contaminant testing, and an initial GMP compliance audit.

So the banned-substance screen is lot-by-lot; the label-claim and contaminant checks are annual; the manufacturing audit happens at the start. That is a sensible design, but "BSCG certified" does not mean every element was checked on every lot, and the marketing shorthand can blur that.

ConsumerLab: not a certification, and the difference matters

ConsumerLab is frequently grouped with the others. It works differently.

Its core business is a subscription testing and review service. ConsumerLab buys products at retail, tests them, and publishes results to paying subscribers as Approved or Not Approved. The consumer pays; the manufacturer is not involved and, for most reviewed products, there is no mark on the label at all.

Separately, ConsumerLab runs an optional program in which manufacturers can pay for testing and, if the product passes, display a seal. ConsumerLab's own description of that program states: "There is a testing fee paid by the manufacturer and results are proprietary to the manufacturer."

The structural difference: USP, NSF, Informed and BSCG put a mark on a label that a manufacturer sought out and paid for. ConsumerLab's main product is subscriber-funded testing of products it buys itself and publishes on (closer to journalism than certification), with a smaller paid seal program alongside.

Neither model is inherently better. They answer different questions. A manufacturer-paid mark tells you a company chose to be examined. An independent retail-purchase test tells you what was in a bottle somebody bought off a shelf.

What none of them do

The same gap shows up in every case:

None of them tests whether the product works. Not one of these programs evaluates efficacy. A certified product with a useless ingredient is a certified product with a useless ingredient.

None of them is an FDA program. See above.

None of them covers every product a company sells. Certification is per-product, sometimes per-lot. A brand that certifies one SKU has not certified the rest of the catalog, and the logo on the website is not the same as the logo on the bottle in your hand.

A mark is not a substitute for the dose being right. A product can be exactly what the label says and still contain an amount too small to be meaningful. Certification confirms the label; it does not evaluate the formula.

Verify the mark, not just the logo

A printed logo is artwork. The major certifiers publish searchable listings of what they have certified (USP and NSF both do), and checking takes under a minute.

This matters for reasons more mundane than outright counterfeiting:

Certification can lapse. A company may have been certified when the packaging was designed and not have renewed. Old stock carries old artwork.

Certification is product-specific, not brand-wide. A company with one certified product may print trust language across its whole site. The listing tells you which SKU was actually examined.

Certification can be facility-specific. "Manufactured in a certified facility" is a different statement from "this product is certified," and the two get used interchangeably in marketing copy.

The scope may not be what the placement implies. A banned-substance certification displayed next to purity language is being asked to carry weight it was not designed to carry.

Find the program's product listing, search the exact product name, and confirm. If it is not in the listing, that is a question worth putting to the company before it is a reason to walk away, since listings lag and errors happen.

When a brand you like has no mark

This comes up constantly and deserves a straight answer rather than a reflex.

The mark tells you a company chose to be examined. It does not tell you that an uncertified company would have failed. Certification costs money and takes time, and it is entirely normal for a small, careful manufacturer to have skipped it while a large company with an indifferent product has paid for it.

What you can do instead is ask the questions the certification would have answered:

• Who manufactures this? If the brand does not make it, which contract manufacturer does, and is that facility audited by anyone?
• What identity testing is done on incoming ingredients, and by what method?
• Is the finished product tested, or only the raw materials? These are different, and a clean ingredient can be contaminated downstream.
• Will you send me the certificate of analysis for the lot number on my bottle?
• Who ran the tests? Was it an in-house lab, the supplier, or an independent one?

The answers matter, and so does the manner of them. A company with a real quality system answers these quickly and specifically, because somebody there already has the documents. A company that treats the questions as unusual is telling you something about how often it gets asked.

How to use this

Check what the mark is, not just that there is one. A sports banned-substance mark and a general quality certification are different products. Read the logo.

Check the specific product. Look it up in the certifier's published listing rather than trusting the artwork.

Match the mark to your reason for caring. Worried about contaminants in a botanical? USP or NSF/ANSI 173. Subject to drug testing? Informed Sport or NSF Certified for Sport, and note the every-batch distinction. Wanting to know whether a category of product generally delivers what it claims? An independent testing publication is more use than any seal.

Do not penalize an uncertified product automatically. Certification costs money and takes time, and plenty of careful small manufacturers have not pursued it. Absence of a mark is absence of information, not evidence of a problem. What you can do is ask the company directly what testing it performs, who performs it, and whether it will share a certificate of analysis for the lot you are buying.

Do not let a mark end your thinking. It is one input. The dose, the form of the ingredient, whether the product suits you, and whether it interacts with your medications are all outside its scope.

The balanced view

These programs are worth having. In a market where no one checks before a product reaches a shelf, a voluntary external check, with facility audits, contaminant screening and label-claim verification behind it, is a real contribution, and companies that pay for one have chosen to be looked at when they did not have to.

But read the mark for what it says. It says something specific about identity, purity, potency and manufacturing. It does not say the product is safe for you, and it does not say the product works. NIH's sentence is the right one to carry with you: these seals do not guarantee that a product is safe or effective.

Used that way, as evidence about the contents and the process and not about the outcome, a certification mark is one of the more informative things on the bottle.