- +1-910-939-3205
- Cary, North Carolina 27518, United States
- hello@fafabiotic.com
It is the number on the label of anything claiming to be alive. It is also more slippery than it looks, and understanding why tells you a great deal about what a company actually knows about its own product.
The short answer
CFU means colony forming unit. It counts colonies that grew, not cells that exist, and those are not the same number. It is a floor rather than a census. And a CFU figure with no time point attached tells you almost nothing.
I have written that any product claiming to contain live organisms should give you a number. This article is about what that number is, and what it is not.
The method is over a century old and pleasingly simple.
1. Take a sample of the product and dilute it, usually in a series of ten-fold steps.
2. Spread a measured volume of a dilution onto a plate of growth medium.
3. Incubate. Anything capable of growing divides, repeatedly, until it forms a visible spot.
4. Count the spots. Multiply back by the dilution. That is your CFU per gram or per millilitre.
Note what is being measured. Not cells. Colonies. The whole method rests on an assumption that one visible spot came from one cell, and that assumption is wrong in two directions.
Lactobacilli often grow in chains. Many bacteria clump. If twenty cells land on the plate as a single clump, they grow into one colony, and they are counted as one.
Which means CFU systematically undercounts. The real number of viable cells present is higher than the figure reported, sometimes considerably.1 The unit is honest about this if you read it literally: a colony forming unit is whatever formed one colony, whether that was one cell or thirty.
The second problem is more interesting.
Cells that have been through stress, and freeze-drying, storage and formulation all count as stress, can enter a state called viable but non-culturable. VBNC cells have intact membranes and are metabolically active. They are alive by any reasonable definition. They simply will not grow into a colony on a plate under those conditions.2
To a plate count, those cells are invisible. They are alive and they register as zero. Some can later resuscitate and become culturable again, which is a strange thing to sit with: a number that can go up without anything being added.
Both errors run the same direction. A CFU count is a floor, not a census. Whatever the number says, there are probably more living cells than that.
Flow cytometry takes a different approach. Rather than asking cells to grow, it passes them single file past a laser and reads them individually, using fluorescent stains that report on membrane integrity, enzyme activity or membrane potential.
There is an international standard for it, ISO 19344 and IDF 232, written for lactic acid bacteria specifically. It reports results as active fluorescent units, or AFU, and it describes three separate staining protocols corresponding to three different definitions of what “viable” means.3
It is faster, it has lower variation between runs, it counts individual cells rather than clumps, and it sees VBNC cells that plating misses.4 On the face of it, better in every way.
Because of what each method is really asking.
CFU asks
“Can this organism grow and multiply?”
A high bar, and the one that matches what people mean by a living organism doing something. Slow, up to five days, and it undercounts.
Flow cytometry asks
“Is this cell’s membrane intact, or its enzymes working?”
Fast, precise, and it counts cells that a plate would miss. But an intact membrane is not proof that a cell can do anything.
Neither is the true answer. They measure different properties and the sensible position is that they are complementary rather than competing.2 A count without a stated method is missing half its meaning.
I publish viability data, and those figures are plate counts. So everything above applies to them.
The direction of the error is worth being clear about. Both problems cause plate counts to understate the number of living cells. My numbers are conservative floors, not best-case figures. I would rather that than the reverse, but you should know that is what you are looking at, and you should apply exactly the same reading to anyone else’s numbers including numbers that look better than mine.
All of the above is detail. This is the part that actually decides whether a label is meaningful.
When was the count taken?
At manufacture. The easiest possible moment, immediately after production, before anything has had time to die. Any number can be true here and mean nothing by the time you open the jar.
At the end of shelf life. The number that costs something to promise, because it requires stability data across the whole period and it constrains what you can print.
At the point of use. The only one that describes what reaches you.
A count with no time point attached is almost always the first kind. If it were the second or third, saying so would be free, and it would be a selling point. Silence here is informative.
✗ The myth
“A bigger CFU number means a better product.”
✓ The fact
Two numbers are only comparable if they were taken by the same method at the same point in time. Usually you are told neither.
✓ Is there a number at all?
✓ At what point in time does it apply?
✓ By what method, plate count or flow cytometry?
✓ Per what? Per gram, per millilitre, or per full application, which are very different quantities.
Key takeaways
✓ CFU means colony forming unit. It counts colonies, not cells.
✓ Chains and clumps grow into one colony, so the method undercounts.
✓ Viable but non-culturable cells are alive and register as zero.
✓ Both errors point the same way, so a CFU figure is a floor.
✓ Flow cytometry under ISO 19344 reports AFU instead, counts individual cells, and sees what plating misses. The two methods answer different questions.
✓ The time point matters more than the number. At manufacture is easy. At end of shelf life costs something to promise.
References
Links go to the peer-reviewed record. You are welcome to check my work.
1. Chiron C, Tompkins TA, Burguière P. Flow cytometry: a versatile technology for specific quantification and viability assessment of micro-organisms in multistrain probiotic products. J Appl Microbiol. 2018. doi:10.1111/jam.13666
2. Probiotic viability reconsidered: integrating VBNC resuscitation and culture-independent methods for accurate probiotic enumeration. Microorganisms. 2025;13(11):2479. mdpi.com
3. ISO 19344:2015 | IDF 232:2015. Milk and milk products. Starter cultures, probiotics and fermented products. Quantification of lactic acid bacteria by flow cytometry. Defines active fluorescent units (AFU) and three staining protocols. iso.org
4. Sudha MR, et al. Rapid assessment of viable but non-culturable Bacillus coagulans MTCC 5856 in commercial formulations using flow cytometry. PLoS One. 2018;13(2):e0192836. doi:10.1371/journal.pone.0192836
5. Rezaei F, Rollin JA. A microbial formulation perspective on probiotic skincare: viability, challenges, and current approaches to maintain probiotic viability. Biotechnol Bioeng. 2026. doi:10.1002/bit.70268
Still have a question?
Submit it here, and it may become the next Ask the Scientist article.
About the author
Dr. Farzaneh Rezaei, PhD, Microbial Formulation Scientist
Founder of Fafabiotic, with two decades developing microbial technologies. She writes Ask the Scientist to explain the science behind probiotic skincare without the marketing layer.
Educational content only. This article is not medical advice and is not intended to diagnose, treat, cure or prevent any disease.
