Independent testing

When Labels And Capsules Disagree: What DNA Testing Finds In Probiotic Products

Two peer-reviewed sources cited on this site's own ingredients page look at probiotic labels from the evidence side: DNA-based authentication methods capable of identifying what is actually in a capsule, and a laboratory audit finding poor correlation between commercial probiotic labels and what testing measures. This article summarises what independent testing, not label-reading, has found.

By SodaTide Team · Last updated September 30, 2026 · 11 min read · Independent testing · Reviewed by Dr. Wendy Scinta, MD

Two different ways to question a probiotic label

There are two genuinely different ways to raise a concern about a probiotic label, and this site has now published articles taking each approach. One approach argues from a standard: here is what a responsible label should print, according to the scientific bodies that study this category, and here is what this label actually prints instead. This site's article on what the SodaTide probiotic blend does not tell you takes that approach, working from the International Society of Sports Nutrition's position stand on what a probiotic product should disclose.

The other approach argues from evidence: independent researchers have physically tested commercial probiotic products against their own labels, using laboratory methods capable of identifying exactly which organisms are present and in what quantity, and here is what they found across the industry when they did. This article takes that second approach. It is not about what SodaTide's label should say. It is about what happens, industry-wide, when someone checks what a probiotic label does say against what is actually inside the bottle.

How DNA-based testing actually identifies what's in a capsule

Shehata and Newmaster, reviewing the state of DNA-based methods in probiotic authentication, describe a set of laboratory techniques capable of identifying which bacterial species and strains are genuinely present in a finished product, distinguishing them from close relatives that might share a genus name but behave very differently, and in some cases quantifying how much of each organism is actually there. This is a meaningfully different kind of check than reading a label. A label is a claim, printed by the company that benefits from the claim being believed. A DNA-based test is a direct measurement of the product itself, run by a party with no stake in which answer it produces.

The existence of this authentication literature is itself informative. Researchers do not build and refine laboratory methods to solve problems nobody has. DNA-based authentication techniques for probiotics exist, and have been refined across multiple papers over several years, specifically because label claims and product contents have not always matched closely enough for a simpler check to settle the question.

What one laboratory audit actually found

Kesavelu and colleagues, publishing in Cureus in 2020, ran a laboratory audit of commercial probiotic products sold in India, comparing what each product's label claimed against what laboratory testing actually measured for identity, viability, and colony count. The finding, stated plainly in the paper's own title, is poor correlation between label claims and measured identity, viability and counts.

Read carefully, that is three separate failure points bundled into one audit, and each one is worth naming individually because they represent different ways a probiotic product can fall short of its own label. Identity failure means the organism a test actually finds is not the organism the label names — a different species or strain entirely, not merely an under-dosed version of the right one. Viability failure means the organisms present are not alive, or not alive in numbers that could plausibly deliver a probiotic effect, regardless of what species they nominally belong to; a probiotic that has died in the bottle is, functionally, no longer a probiotic. Count failure means the colony-forming-unit number printed on the label does not match what a test actually measures, in either direction, though a shortfall relative to the labelled figure is the more common and more consequential direction in practice. A single product can fail on one of these axes, two, or all three at once, and a laboratory audit designed to check all three, the way this one did, gives a more complete picture than a test checking only one.

Framing based on the three outcomes a laboratory audit of this kind is designed to check, per Kesavelu 2020.

Why label-to-content gaps happen at all

None of this requires assuming bad faith on the part of every manufacturer whose product fails an audit, and it is worth being fair about that before going further. Live bacterial organisms are harder to keep stable, in an accurate and verifiable state, than a synthetic vitamin or mineral compound. Viability can decline over a product's shelf life even when the label was printed honestly at the point of manufacture, which is exactly why the probiotic field's own standards — already covered in this site's separate article on what the blend does not tell you — call for a count at the end of shelf life rather than at the moment of production. A count that was accurate on day one and has simply declined by the time a bottle reaches a shelf is a different, more sympathetic failure than a count that was never accurate at all.

Manufacturing conditions matter too: temperature during storage and transport, moisture exposure, and time all affect how many organisms in a probiotic product remain alive and correctly identified by the time an independent lab, or a buyer's own digestive tract, encounters them. None of this is a defence of any specific product. It is context for why an industry-wide audit finding poor correlation between label and content is plausible on its face, rather than an implausible claim that would need extraordinary evidence to credit.

How this fits the wider probiotic category

Probiotics sit in an unusual regulatory position in most markets, including the United States: sold as dietary supplements, they are not required to clear the pre-market efficacy and manufacturing review that a pharmaceutical product undergoes before reaching a shelf. That lighter regulatory load is part of what makes independent, investigator-initiated audits like Kesavelu's valuable in the first place — they supply a check that the approval process for this product category does not itself guarantee. Shehata and Newmaster's review of DNA-based authentication methods exists in that same gap, offering manufacturers, regulators and independent researchers a technical means of running the check that the category's own approval pathway does not require by default.

None of that is unique to probiotics as a supplement category, and none of it is an argument that every probiotic product is suspect. It is the background condition that makes this entire body of audit literature necessary, and it is the reason a label's own claim, printed by the party selling the product, is not by itself the same thing as independent proof.

What this means for a label that names no strains at all

Here is where this article's evidence-based approach and this site's standards-based article on the same blend converge on the same practical conclusion, by two different routes. If independent laboratory audits find that even labels which do name a specific genus, species, strain and colony count do not reliably match what testing finds, then a label that declines to name any of those things in the first place has not merely skipped a formality — it has removed the one thing that would let a reader, or a lab, check it at all.

A specific label can be tested against its own specific claim, as the Kesavelu audit did for each product it examined, and can be found accurate or inaccurate. A blend with no genus, species, strain, or count printed anywhere makes that test impossible to run in the first place, because there is no specific claim to check. That is not a smaller problem than a label that makes a false claim. In an important sense, it is a more complete one: a false claim can eventually be caught and corrected by exactly the kind of audit this article describes; a blend that names nothing has pre-emptively made itself un-auditable by the same method.

Reading this alongside the blend's own silence

What a buyer can actually do with this information

This article is not a claim that SodaTide's own probiotic blend has failed, or would fail, a DNA-based audit, and it would be a fabrication to imply otherwise — no such test of this specific product is cited here because this site is not aware of one having been published. What this article establishes is the wider evidentiary context a reader should hold in mind when a probiotic label, any probiotic label, asks to be trusted on its printed claims alone: independent testing of the category has repeatedly found gaps between what labels say and what products contain, using methods capable of checking identity, viability and count all at once.

The practical upshot is a reasonable, calibrated scepticism rather than a blanket dismissal. A label that prints genus, species, strain and an end-of-shelf-life count has at least made a checkable claim, and checkable claims can be verified by exactly the kind of DNA-based methods this article describes. A label that prints none of those things has made a claim that cannot be checked by anyone, including the seller, which is a strictly worse epistemic position for a buyer to be in — not because it proves the product is bad, but because it removes the option of finding out.

References