A "live cultures" claim on a jar of kimchi, miso, or yogurt tells you almost nothing on its own. It does not name a strain, does not give a dose, and does not tell you whether that dose survived the shelf. A genuine probiotic claim requires a characterized strain, a stated CFU count, and evidence tying that dose to a benefit. Most fermented foods are honest, live-culture foods that never make that claim at all, which is fine, as long as you are not paying a premium for a promise the label was never making.
1. What "Live and Active Cultures" Legally Means
In the United States, the phrase most shoppers actually see is not a government-mandated claim. It is a voluntary seal. The National Yogurt Association (NYA) created the Live & Active Cultures seal in the 1980s, and a manufacturer that wants to display it has to meet one specific bar: the product must contain at least 100 million cultures per gram at the time of manufacture. Refrigerated products carry the seal at that threshold; frozen products qualify at a lower bar because freezing kills off a larger share of the population.
Two things about that standard are worth sitting with. First, it is a manufacturing-date requirement, not a point-of-sale requirement. Nothing about the seal guarantees that the same count is present when the jar reaches your cart weeks later, and shelf life reliably erodes it. Second, participation is voluntary. A product without the seal is not necessarily lower quality; it may simply belong to a brand that never applied, or a category, like most fermented vegetables, where the seal was never built for that product type in the first place.
Outside the NYA seal, "contains live cultures," "raw," and "unpasteurized" are largely unregulated descriptive phrases. They can be accurate and still tell you nothing quantitative. A jar that says "live cultures" and a jar that says "100 million CFU per gram" are making very different strength of claim, and only one of them can actually be checked against a number.
If a label uses the phrase "live cultures" but never prints a number, a strain name, or a "best by" context for that number, treat it as a general statement of fact about the fermentation process, not a measurable health claim.
2. Why a Fermented Food Is Not Automatically a Probiotic
This is the gap that causes the most confusion, and it is not a marketing trick so much as a definitional one. In 2014, a panel convened by the International Scientific Association for Probiotics and Prebiotics (ISAPP) published a consensus statement setting out what the word "probiotic" is supposed to mean in a scientific context: live microorganisms that, when administered in adequate amounts, confer a health benefit on the host.
Read carefully, that definition has three separate requirements stacked inside it, and a food has to clear all three to earn the word honestly.
- A characterized strain. Not just a genus like "Lactobacillus," and not even just a species like "Lactobacillus plantarum," but a specific, identified strain with its own designation, the way a drug has a specific molecule rather than a chemical family.
- A defined, adequate dose. A quantity known to be sufficient to survive digestion and do something, not a trace amount that happened to be present.
- Demonstrated evidence of benefit. A study, ideally more than one, showing that strain at that dose actually did something measurable in humans.
A jar of naturally fermented kimchi genuinely contains live bacteria, mostly from the Leuconostoc and Lactobacillus genera involved in the fermentation succession. That is not nothing. But almost no commercial kimchi label identifies the specific strains present, states a CFU dose, or cites a clinical trial run on that exact product. It is, correctly described, a live-culture fermented food. Calling it "a probiotic" without those three elements is technically imprecise, even when the underlying microbiology is completely real.
This distinction is not a reason to be suspicious of fermented foods. It is a reason to hold two separate ideas at once: fermented foods are generally worth eating for reasons that go beyond any single bacterial strain, including organic acids, altered nutrient availability, and the fermentation process itself, while a specific, dosed, evidence-backed probiotic effect is a narrower and rarer claim that only some products, mostly supplements and some dairy products, are actually positioned to make.
3. Pasteurized vs Unpasteurized: Kimchi, Sauerkraut, and Miso
Heat is the single biggest variable standing between "this jar has live cultures" and "this jar used to." Pasteurization, heating a food to a temperature that kills essentially all vegetative microorganisms, is what allows a fermented product to sit safely at room temperature on a regular grocery shelf indefinitely. An unpasteurized ferment, left warm, would keep fermenting, keep producing gas, and eventually push against or pop its own lid.
That mechanical reality is actually a useful, if imperfect, shortcut for shoppers:
| Signal | Likely pasteurized | Likely unpasteurized |
|---|---|---|
| Storage location | Shelf-stable, room-temperature aisle | Refrigerated case |
| Packaging behavior | Sealed jar, no venting needed | Sometimes vents gas, "burps" on opening |
| Label language | Often says "pasteurized" or omits culture claims entirely | Often says "raw," "live cultures," or "unpasteurized" |
| Sell-by window | Long, sometimes a year or more | Shorter, and quality keeps shifting after purchase |
None of these signals is a guarantee by itself, storage practices vary by store, and some refrigerated products are pasteurized anyway for consistency reasons. But taken together, they are a reasonable field check when the label does not spell it out directly. The same logic applies across the major lacto-fermented vegetables, including sauerkraut, where shelf-stable canned versions have almost always been heat-processed while refrigerated bagged versions usually have not.
Miso is a slightly different case, because most miso, pasteurized or not, is sold shelf-stable or refrigerated as a thick paste rather than a jar of visible liquid, so the venting cue does not apply. Some miso producers pasteurize specifically to lock in a consistent flavor profile and halt further fermentation; others sell unpasteurized paste that continues to slowly develop in the tub. Neither is more "authentic" by default, they are different production choices, but only one of them is still delivering live cultures by the time it reaches a spoon. For the safety chemistry behind why these processing choices exist at all, the site's fermentation safety guide covers the pH and pathogen-control side of it in more depth.
4. Why the Number on the Label Is Not the Number You Eat
Even a product that does everything right, names a strain, prints a CFU count, and was genuinely unpasteurized at the time of packing, is still describing a population that only moves in one direction after that measurement is taken: down. Live microorganisms are, definitionally, living things with a shelf life of their own, and CFU counts decline in storage as some fraction of the population dies off over time.
The rate of that decline depends on several variables working together: the specific organism and strain (some are hardier than others), the storage temperature (refrigeration slows die-off considerably; room temperature and heat cycling accelerate it), the packaging (oxygen exposure is generally harder on strict anaerobes), and simple elapsed time. A product tested and labeled the week it was made is describing a population that was already smaller by the time it reached a store shelf, and smaller again by the "best by" date, and smaller still if it sat in a warm car on the way home.
Buy live-culture products as close to their production date as you reasonably can, keep them consistently refrigerated once opened if refrigeration is called for, and treat any printed CFU number as a starting point that was true on day one, not a promise about the count in the container in your hand right now.
5. Cooking Miso and Gochujang: Losing the Cultures, Keeping the Flavor
This is where a lot of otherwise careful home cooks lose the thread. Fermentation microorganisms are not heat-stable much past roughly 60 to 70 degrees Celsius, and a simmering soup runs well above that threshold. Stirring a spoonful of miso into hot dashi to make miso soup, simmering kimchi into kimchi-jjigae, or melting gochujang into a simmering braise all reliably kill the live cultures those ingredients were carrying.
That is not a reason to avoid cooking with them, and it is not a hidden flaw in the ingredient. Fermentation does two separate jobs, and only one of them requires the organisms to still be alive when you eat the food. The first job is live microbial exposure, which cooking ends. The second job is chemical transformation that already happened during fermentation and does not undo itself with heat: proteins broken down into free amino acids, starches converted into sugars, and the buildup of glutamate and other compounds responsible for umami. Miso's savoriness, gochujang's depth, and kimchi's tang are the residue of finished fermentation chemistry, not a live process that heat interrupts mid-stream.
In practice this means a useful, simple rule: if the goal is live-culture exposure, eat the ferment cold or only lightly warmed and add it at the very end, off heat or close to it. If the goal is flavor, cook it in freely, the umami and depth are staying regardless of what happens to the bacteria.
6. What the Label Can and Cannot Tell You, and How to Check a Strain
Put together, a fermented food label can reliably tell you a few things and cannot tell you several others, and it is worth being precise about the difference before deciding what to trust.
What it can tell you: whether the product claims to be pasteurized or not, whether it displays a recognized seal like the NYA Live & Active Cultures mark and the manufacturing-date threshold that seal implies, and, on the better labels, a specific strain designation and a CFU count as measured at a stated point in time.
What it usually cannot tell you: the CFU count at the moment you actually eat it, whether the specific strain present has any published clinical evidence behind it at all, or whether the amount surviving digestion is anywhere near the dose used in whatever study a marketing claim is loosely gesturing toward.
The single highest-leverage check a shopper can do takes under a minute. If a label names a strain, it will typically look like a genus, a species, and a code: Lactobacillus rhamnosus GG, Bifidobacterium animalis subsp. lactis BB-12, Lacticaseibacillus paracasei Lpc-37. Copy that exact strain designation, including the code, into PubMed, the free biomedical research database maintained by the U.S. National Library of Medicine. A genuinely well-studied strain will surface named clinical trials. A label that lists only a bare genus, "contains Lactobacillus," with no species and no code, has not actually given you anything specific enough to search, which is itself useful information: there is no way to verify a claim that was never made at a checkable level of detail.
7. The 60-Second Label Checklist
Frequently Asked Questions
Does "contains live cultures" on a label mean the food is a probiotic?
Not on its own. "Live cultures" only tells you that some viable microorganisms were present at some point. The scientific definition of probiotic, set out by the International Scientific Association for Probiotics and Prebiotics (ISAPP), requires a characterized strain, a defined dose, and evidence that dose confers a specific health benefit. A jar of kimchi carries live bacteria, but almost never carries strain identification, a CFU count, or a dose-linked health claim, so most fermented foods on a shelf are correctly described as live-culture foods, not verified probiotics.
How do I know if my kimchi, sauerkraut, or miso is pasteurized?
Check the label for the word "pasteurized" and check storage placement. Shelf-stable jars sitting at room temperature in a regular grocery aisle have almost always been heat-treated, since an untreated ferment kept warm would keep fermenting and eventually pop its lid or spoil. Refrigerated kimchi, sauerkraut, and miso sold cold are far more likely to be unpasteurized and still fermenting slowly, though a cold shelf is a strong clue rather than a guarantee. When the label does not say either way, the fermented-foods aisle staff or the manufacturer's website is the reliable next check.
Does cooking kimchi, miso, or gochujang destroy the live cultures?
Yes. Lactic acid bacteria and other fermentation microbes are not heat-tolerant past roughly 60 to 70 degrees Celsius, and a simmering soup or stew runs well above that. Stirring miso into hot broth, cooking kimchi into kimchi-jjigae, or dissolving gochujang into a simmering sauce kills essentially all the live cultures those ingredients carried. The flavor compounds built during fermentation, the amino acids, organic acids, and glutamate that create umami, survive cooking intact. You lose the live-culture exposure but keep the flavor payoff, which is a real trade, not a loss across the board.
Why does CFU count matter and does it stay the same in the jar?
CFU, colony-forming units, is the standard way of counting viable microorganisms per gram or per serving. It matters because a trace amount of a live organism is a different exposure than a well-populated dose. Any CFU figure on a label describes the count at the time of manufacture, not the count you eat, and that number declines over shelf life as some organisms die off during storage, faster in warm conditions and slower under refrigeration. A product date-stamped months ago has a lower live count than the number printed on the label, even if nothing on the label has changed.
How can I check whether a specific probiotic strain has real evidence behind it?
Look for a strain designation on the label, a genus, species, and an alphanumeric code such as Lactobacillus rhamnosus GG or Bifidobacterium animalis subsp. lactis BB-12, rather than just a genus name. Search that exact strain designation on PubMed, the free biomedical literature database run by the National Library of Medicine. A well-studied strain will return clinical trials naming it specifically. If the label lists only a genus like "Lactobacillus" with no species or strain code, there is no specific strain to search, and the claim cannot be verified against the literature at all.