CFU (Colony-Forming Units): What the Number Means, "At Expiry" vs "At Manufacture" and Why More Is Not Always Better
What CFU (colony-forming units) means on a probiotic label, why the time of the count matters, why trial amounts belong to a strain, and why more is not always better.
Content last updated . This page has not been medically reviewed.
This page explains the number. It does not give a recommended amount; a clinician or pharmacist can help with a specific product.
How the sources define the term
In a 12-month storage study, viable microbial counts were measured by culture-based enumeration. [4] Health Canada's archived 2010 notice for natural health product licence applications said applicants must present the viable counts of the probiotic microorganisms in the product. [5] The FAO/WHO guidelines, as summarized by ISAPP, say labels should show the minimum viable numbers of each strain at the end of the shelf life, typically as CFU. [1]"At expiry" versus "at manufacture"
These are two different promises.- At manufacture is the count when the product was made.
- At end of shelf life is the count that should remain by the expiry date.
ISAPP says most probiotic strains lose cell count over shelf life, some more than others, and that a count given "at time of manufacture" is inadequate if you buy the product close to the end of its shelf life. [1] In ISAPP's survey summary, about as many products as lacked a strain designation also did not give a measure of potency through the end of shelf life. [1] The archived Health Canada notice required that the label-claim number of viable organisms be present at the end of the expiry period. [5] That notice is archived and not updated, and we did not read other countries' rules.
What research found about counts
| Question | What the source found | Strength of evidence |
|---|---|---|
| Does a stated count hold through shelf life? | In a 12-month test of commercial multi-strain supplements, compliance with label-declared counts declined toward the end of minimum durability, and some products missed declared values whatever the storage. [4] | One laboratory study; not a market ranking |
| Do products match their labels? | A laboratory study of locally marketed products (authors in Pakistan) reported almost none met the label claim and 11% had no viable cells. [6] A review said products are frequently not compliant for strain and viable cell number. [7] | Mixed, from different markets and methods |
| Does a higher dose work better? | A Cochrane review of children's antibiotic-associated diarrhea found a credible subgroup effect for at least 5 billion CFU a day versus less. [8] The authors said the benefit of high-dose probiotics needs confirming in a large trial, and noted a diversity of probiotic agents used. [8] | Moderate certainty for that pooled subgroup; does not transfer to other strains or conditions |
| Does the total count tell you what each strain contributes? | ISAPP says in a multi-strain product you do not know whether equal amounts of all strains are included. [1] | Expert commentary |
Study amounts are not label advice
Trial doses are tied to the strain, product, population and outcome. Two examples from the AGA technical review, both in adults with C. difficile infection also receiving antibiotics: Saccharomyces boulardii at 3 x 10^10 CFU a day may have increased the cessation of diarrhea (low certainty), while Lactobacillus plantarum 299v at 5 x 10^10 CFU a day may not have (very uncertain evidence). [9] These are study amounts. They show that a large number does not guarantee a result, and they are not advice. C. difficile infection is treated by clinicians; see Probiotics and C. difficile.A systematic review of 228 trials warned that pooling different probiotics gives misleading conclusions. [10] The AGA noted that there are no guidelines for reporting product testing in trials, so results for the same strain from different manufacturers are hard to compare. [9]
Related pages
See Probiotic strains and labels, How to read a probiotic label and Probiotic storage.Frequently asked questions
How many CFU should a probiotic have?
No universal number exists in the sources we read. NCCIH says how much people would have to take is unknown. [2] A clinician or pharmacist can discuss a specific product.
What does "billion CFU" mean?
It means a count of viable organisms, in billions. It does not say how many of them reach the gut or whether the strain has been studied for your question. [3]
What is the difference between CFU at expiry and at manufacture?
At expiry is the count that should remain by the expiry date. At manufacture is the starting count, which ISAPP calls inadequate if you buy close to the end of shelf life. [1]
Is a higher CFU always better?
No. A Cochrane subgroup finding favored at least 5 billion CFU a day in children's antibiotic-associated diarrhea, but it pooled many different probiotics and needs confirmation. [8] It does not apply to every strain.
Are the CFU numbers in studies a dose I can use?
No. They are study amounts for a specific strain and group. [9] Ask a clinician or pharmacist.
More details: evidence, limits and cautions
The details behind this page: cautions, limits of the evidence and where sources disagree. Open any note to read it.
Safety note
Limits and uncertainty
- A CFU count does not show that bacteria survive stomach acid or reach the gut; the NHS says we cannot always be sure of that. [3]
- Counting methods differ: a review of authenticity says a plethora of methods exist, each with advantages and disadvantages, and that they should be standardized. [7]
- "Proper evidence not found": we searched PubMed and the guidance pages above for a trial-proven minimum CFU that applies across strains, and found none. NCCIH's statement that the amount is unknown points the same way. [2]
References
- Sanders ME. Decoding a Probiotic Product Label. International Scientific Association for Probiotics and Prebiotics (ISAPP), 2022, updated 21 June 2025. Read 2026-10-22.
- NCCIH (US National Center for Complementary and Integrative Health). Probiotics: Usefulness and Safety. Page shows Last Updated August 2019. Read 2026-10-22.
- NHS (UK). Probiotics. Page last reviewed 24 November 2022. Read 2026-10-22.
- Brousilova T et al. Storage temperature as a determinant of probiotic long-term stability in commercial dietary supplements. Front Nutr 2026. PMID 42824401. Read 2026-10-22.
- Health Canada. ARCHIVED - Notice - Product Licence Applications for Natural Health Products Containing Probiotics. Notice dated January 2010; page archived by Health Canada. Read 2026-10-22.
- Aziz G, Zaidi A, Tariq M. Compositional Quality and Possible Gastrointestinal Performance of Marketed Probiotic Supplements. Probiotics Antimicrob Proteins 2022. PMID 35199309. Read 2026-10-22.
- Authenticity of probiotic foods and dietary supplements: A pivotal issue to address. Crit Rev Food Sci Nutr 2022 (erratum 2023). PMID 33819118. Read 2026-10-22.
- Guo Q et al. Probiotics for the prevention of pediatric antibiotic-associated diarrhea (Cochrane review). Cochrane Database Syst Rev 2019. PMID 31039287. Read 2026-10-22.
- Preidis GA et al. AGA Technical Review on the Role of Probiotics in the Management of Gastrointestinal Disorders. Gastroenterology 2020 (full text read via PMC8018518). PMID 32531292. Read 2026-10-22.
- McFarland LV, Evans CT, Goldstein EJC. Strain-Specificity and Disease-Specificity of Probiotic Efficacy: A Systematic Review and Meta-Analysis. Front Med 2018. PMID 29868585. Read 2026-10-22.