CFU vs AFU: How to Read a Probiotic Label

Medically reviewed and written by Dr Zeeshan Afzal, MBBS — Medical Officer and Health Content Lead at Welzo. Last updated: July 2026.

If you have compared two probiotic bottles in a UK pharmacy recently, you have probably noticed that the numbers no longer match up. One product promises 30 billion CFU. Another claims 53 billion AFU. A third quietly says "10 billion live cultures" and leaves you to guess. The CFU vs AFU question is not marketing noise — the two units are produced by genuinely different laboratory methods, and they answer slightly different questions about what is actually alive inside the capsule.

This guide explains what each unit measures, why AFU figures usually look larger, and how to read the rest of a probiotic label so the number is put in context. If you are still deciding what to buy, our gut health range and probiotics collection are a good starting point, alongside our practical guide on how to choose a probiotic and our roundup of the best probiotics in the UK. Products measured in different units — such as Akkermansia muciniphila — are covered later in this article, and non-live gut supplements such as modified citrus pectin powder, Welzo Ultra Purity Berberine and Welzo Ultra Purity TUDCA are not measured in CFU or AFU at all, because they contain no live organisms.

Key takeaways

• CFU (colony forming units) counts bacteria that can grow into a visible colony on an agar plate. It is the traditional unit and the one used in almost every published clinical trial.
• AFU (active fluorescent units) counts bacteria with an intact, functioning cell membrane using flow cytometry, under the international method ISO 19344 | IDF 232.
• AFU figures are frequently higher than CFU figures for the same material, because flow cytometry also detects viable-but-non-culturable cells and counts clumped cells individually.
• There is no fixed conversion factor between CFU and AFU. You cannot compare a 50 billion AFU product with a 50 billion CFU product as though the numbers were interchangeable.
• The unit is only one line on the label. Strain designation, dose per serving, and whether potency is guaranteed to the end of shelf life matter at least as much.

Table of contents

1. What does CFU mean on a probiotic label?
2. The real limitations of the plate count
3. What does AFU mean on a probiotic label?
4. Where AFU falls short
5. CFU vs AFU side by side
6. Why AFU numbers are usually bigger
7. Why the unit matters for clinical evidence
8. How to read a probiotic label, line by line
9. What UK labelling rules actually allow
10. Is a bigger number a better probiotic?
11. Yeasts, spores and Akkermansia: the special cases
12. A practical label checklist
13. When to speak to a healthcare professional
14. Frequently asked questions
15. References

What does CFU mean on a probiotic label?

CFU stands for colony forming unit. It is a measure of how many microorganisms in a sample are capable of dividing repeatedly until they form a colony visible to the naked eye. It has been the standard currency of probiotic labelling for decades, and it is the unit used in the internationally accepted definition of a probiotic as live microorganisms that confer a health benefit when given in adequate amounts.

How a CFU count is actually performed

The technique is deliberately simple, which is part of its appeal:

1. A weighed sample of the finished product is dissolved and diluted in a series of steps, often down to one part in a million or less.
2. A measured volume of each dilution is spread onto — or mixed into — a nutrient agar plate chosen to suit the organism.
3. Plates are incubated for a set time and temperature, typically 48 to 72 hours, often anaerobically for bifidobacteria.
4. Plates carrying roughly 30 to 300 colonies are counted, and the count is multiplied back up by the dilution factor to give CFU per gram or per capsule.

What a CFU count genuinely proves

A colony is hard evidence. It shows that a cell was not merely intact but was metabolically capable of reproducing under laboratory conditions. That is a meaningful threshold, and it is the reason regulators, pharmacists and trial designers have stayed loyal to it. When a study reports that a strain worked at ten billion per day, that dose was almost certainly established by plate count. Our article on whether probiotics work looks at that evidence base in more detail.

The real limitations of the plate count

Plate counting is trusted, but it is not precise, and understanding why is the whole basis of the CFU vs AFU debate.

One colony does not always mean one cell

Lactobacilli grow in chains. Bifidobacteria form branched, Y-shaped clusters. If eight cells are stuck together when they land on the agar, they produce a single colony and are recorded as one CFU. The plate systematically undercounts organisms that like to clump.

Viable but non-culturable cells are invisible

Freeze-drying, compression into tablets, exposure to moisture and simple ageing all push bacteria into a stressed state in which they remain alive and metabolically active but temporarily refuse to grow on standard media. Microbiologists call these viable-but-non-culturable, or VBNC, cells. To a plate count they may as well not exist. Research using fluorescence in situ hybridisation combined with flow cytometry has directly challenged the "gold standard" status of plate counting on precisely this basis, showing it is predisposed to underestimate true viable cell numbers.

Multi-strain blends are hard to separate

A ten-strain formula is a genuine problem for agar. Some strains outgrow and mask others. Selective media are imperfect. There is a well-recognised shortage of robust methods to quantify individual strains once they are blended, which is one reason so many labels give a single total rather than a per-strain breakdown. If you are weighing up complexity against simplicity, see single vs multi-strain probiotics.

It is slow and variable

Results take days rather than hours. Because so much depends on operator technique, media batch and incubation conditions, reproducibility between laboratories is only moderate. One 2023 comparison found a relative standard deviation of roughly 13% for plate counts against about 7% for flow cytometry on the same material.

Label accuracy has been repeatedly questioned

The consequences show up in market surveys. In an analysis of 28 commercially available probiotic products published in Annals of Microbiology, around 46% of products returned CFU counts below their labelled values. A 2025 UK study of twelve commercial probiotics found that eleven of twelve fell within a practical acceptance range of ±0.5 log CFU of their declared value, but one human product contained no detectable viable bacteria at all. Some of that gap is genuine underfilling or die-off. Some of it is the measurement method itself.

What does AFU mean on a probiotic label?

AFU stands for active fluorescent unit. Instead of asking "will this cell grow?", flow cytometry asks "is this cell's membrane intact?" — a different and, in some respects, broader definition of being alive.

The method behind the number

AFU is defined by an international standard — ISO 19344 | IDF 232, published in December 2015 by the International Organization for Standardization jointly with the International Dairy Federation, and reviewed and confirmed as current in 2021. The standard specifies quantification of active and total lactic acid bacteria and probiotic strains by flow cytometry.

In practice the sample is stained with two dyes. A cell-permeant nucleic acid stain enters every cell and makes it fluoresce green. Propidium iodide can only cross a damaged membrane, and where it enters it fluoresces red and suppresses the green signal. The suspension is then pushed single file past a laser, and each cell is classified by its fluorescence:

• AFU — active fluorescent units: intact membranes, bright green. These are the cells counted as "live".
• n-AFU — non-active fluorescent units: broken membranes, red. Counted as dead.
• TFU — total fluorescent units: the sum of both, i.e. every cell present, alive or not.

Where flow cytometry clearly wins

• Speed. Results are available in around two hours rather than up to 96 hours for a plate count.
• Precision. The interlaboratory validation study behind ISO 19344 reported a repeatability limit of 0.06 and a reproducibility limit of 0.45 for log₁₀ AFU per gram in non-yoghurt matrices.
• Recovery. In a 2023 BMC Microbiology study of multi-species blends, flow cytometry recovered 101.8% ± 6.95% of the known viable cell input, compared with 81.37% ± 16.03% for plate counting.
• Stability data. A 24-month pilot study using ISO 19344 tracked freeze-dried capsules at 25°C and 60% relative humidity and recorded losses of just 0.12 log₁₀ for a single-strain product, 0.16 for a two-strain product and 0.26 for a multi-strain product.
• Validated on real products. The method has been formally validated across three industrial batches for quantifying Lacticaseibacillus rhamnosus GG in a finished formulation — the strain covered in our guide to Lactobacillus rhamnosus GG.

Where AFU falls short

Flow cytometry is not a free upgrade, and it is worth being honest about the trade-offs.

An intact membrane is not the same as a health benefit

A cell that cannot reproduce but still has a sealed membrane will be counted as an AFU. Whether such a cell delivers the same benefit as one that colonises and multiplies is not settled. This is the central scientific criticism of AFU labelling: the burden falls on the manufacturer to demonstrate how membrane integrity relates to biological viability across the full shelf life of that specific strain and formulation.

The method is not species-specific

ISO 19344 itself notes that, because the technique is non-specific, it may quantify bacteria other than those intended when they are present in the sample, which can lead to overestimation. It also does not perform taxonomic differentiation — so it tells you how many intact cells there are, not which organisms they are. Identity has to be confirmed separately, usually by DNA-based methods.

Matrix interference

Clean single-ingredient powders test easily. Microencapsulated products, fortified foods and capsules blended with fibres or botanicals introduce debris that can confuse the detector and require special sample preparation.

The standard was written for dairy

The scope of ISO 19344 covers starter cultures, probiotics and fermented milk products such as yoghurt. It is widely applied beyond dairy, and validation work supports that, but the original remit is narrower than the way the unit is now used commercially.

CFU vs AFU side by side

Why AFU numbers are usually bigger

Three effects push the AFU figure above the CFU figure for identical material.

1. VBNC cells are included. Every stressed-but-intact cell that would never form a colony still registers as an AFU.
2. Clumps are separated. A chain of six lactobacilli is one CFU but six AFU.
3. Membrane integrity degrades more slowly than culturability. As a product ages, cultivability falls away faster than membrane integrity does, so the gap between the two numbers widens over shelf life.

Can you convert AFU to CFU?

No — not reliably. In a freshly manufactured, unstressed culture the two figures can be broadly comparable, but they diverge as the material degrades or as other ingredients are introduced. That divergence depends on the strain, the growth conditions, the drying process and the formulation, so there is no universal multiplier. Any brand that publishes a conversion ratio is describing its own product, not a general rule.

The practical consequence: if two products use different units, you cannot rank them by their headline numbers. Compare CFU with CFU, and AFU with AFU.

Why the unit matters for clinical evidence

This is the point most label guides skip, and it is the most important one.

Essentially the entire clinical literature on probiotics — the trials underpinning any credible claim about a strain — reports dosing in CFU. When a product is labelled only in AFU, you lose the ability to check the dose against the trial that justifies it. A 53 billion AFU capsule may or may not deliver the CFU dose used in the published research on those strains.

This is why the most transparent labels state both. It is also why AFU alone is a weaker signal for a consumer trying to match a product to evidence, even though it may be the more precise laboratory measurement. If you want to trace a strain back to its research, you need the full strain designation — which brings us to the rest of the label. Our pieces on high strength probiotics and how long probiotics take to work both depend on being able to match dose to trial.

How to read a probiotic label, line by line

1. The strain designation

A complete designation has three parts: genus, species and a strain code. For example, Lacticaseibacillus rhamnosus GG. The genus and species tell you the family. The alphanumeric code identifies the exact strain, and effects are strain-specific — two strains of the same species can behave quite differently. Without the code, you cannot find the research.

Note that many genus names changed in 2020. A taxonomic revision published in the International Journal of Systematic and Evolutionary Microbiology split the 261 species then held under Lactobacillus across 25 genera. L. rhamnosus became Lacticaseibacillus rhamnosus; L. plantarum became Lactiplantibacillus plantarum. Both old and new names still circulate on packaging, and neither is a red flag. See our strain guides on Bifidobacterium longum and Lactobacillus plantarum.

2. The number — and the unit next to it

Check whether it says CFU, AFU, or something vaguer like "live cultures" or "active cells". An unspecified unit is the weakest disclosure of the three.

3. "At manufacture" versus "end of shelf life"

This single phrase can change the meaning of the number more than the choice of unit does. A count guaranteed to the end of shelf life is a floor — the minimum you should still receive on the expiry date. A count at time of manufacture is a ceiling that begins falling immediately. Best-practice guidance is that labels should state the minimum viable number of each strain at the end of shelf life. If a label is silent on this, assume it is a manufacturing-date figure.

4. Per capsule or per serving?

A "50 billion" front-of-pack claim sometimes refers to a two- or three-capsule serving. Always read the back panel and divide accordingly.

5. Per strain, or one total?

A single blended total tells you nothing about the ratio. In a ten-strain product, the bulk of the count may come from whichever organism is cheapest to grow rather than the one with the evidence behind it. Per-strain breakdowns are the gold standard, even though they are harder for manufacturers to verify.

6. Storage, format and delivery

Some strains genuinely need refrigeration; many modern freeze-dried and spore formats do not. Water activity is a key driver of die-off, which is why desiccants and blister packs matter. Related reading: do probiotics need refrigeration, shelf-stable probiotics and whether probiotics survive stomach acid.

7. Everything else

Look for a batch or lot number, a visible expiry date, allergen declarations (dairy carriers are common), and whether the brand publishes third-party certificates of analysis. Also check the format suits you — see vegan probiotics, probiotic powders and probiotic drinks.

What UK labelling rules actually allow

British packaging looks different from American packaging for a reason. Under retained EU Regulation 1924/2006, the Department of Health and Social Care treats the word "probiotic" as a health claim in itself, because it implies a relationship between a food and health. Government guidance on nutrition and health claims lists "probiotic" explicitly as an example of a health claim. No probiotic health claim currently sits on the GB Nutrition and Health Claims Register, so the term generally cannot be used on the label of a UK food supplement, and the Advertising Standards Authority has ruled against its use in product names.

That is why UK labels so often read "live cultures", "friendly bacteria" or simply name the strains. It is a regulatory constraint, not evasiveness — and it means you have to work slightly harder to identify what you are buying. It also means UK brands cannot legally tell you a product treats a condition; for that, see the NHS overview of probiotics linked in the references.

Is a bigger number a better probiotic?

Not by itself. Three things matter more than the headline figure:

• Strain-condition match. A strain studied for antibiotic-associated diarrhoea is not automatically useful for bloating. Effects are strain-specific.
• Dose matched to trial. The relevant question is not "is this a big number?" but "does this deliver the dose used in the research for this strain?"
• Survival to the gut and to expiry. A guaranteed end-of-shelf-life count at a modest level beats an unguaranteed enormous one.

Very high counts do have their place — most obviously alongside or after antibiotics, covered in probiotics after antibiotics. For everyday use, dietary diversity does a lot of the work too; see microbiome diversity and prebiotic vs probiotic.

Yeasts, spores and Akkermansia: the special cases

Saccharomyces boulardii

This is a yeast, not a bacterium, and it is often labelled in milligrams rather than CFU because a defined weight of the freeze-dried preparation corresponds to a known cell count. Where CFU is given, both units may appear. More in our guide to Saccharomyces boulardii.

Spore-forming Bacillus strains

Spores are dormant by design and survive stomach acid without protection, but their dormancy is exactly what makes enumeration awkward — a spore that has not germinated may not form a colony on a given medium. This is one area where flow cytometry-based counting is genuinely useful. See spore-based probiotics and Bacillus coagulans.

Akkermansia muciniphila

Akkermansia is where the unit debate has become most heated. It is a strictly anaerobic, notoriously difficult organism to culture, so many products are labelled in AFU or TFU rather than CFU — and TFU includes dead cells, which is a materially different claim again. Some formulations use deliberately pasteurised (non-live) cells, for which live-cell units are not the right measure at all. Our Akkermansia supplement page sets out its own specification, and the background is covered in Akkermansia vs probiotics and pasteurised Akkermansia.

Non-live gut supplements

Products without living organisms — fibres, polyphenols, bile acids, botanicals — are dosed in milligrams or grams. There is no CFU or AFU figure to look for on citrus pectin powder, berberine or TUDCA, and their absence is not a quality concern.

A practical label checklist

Two further practical points: probiotics are generally well tolerated, but mild wind and bloating in the first week or two are common — see probiotic side effects and probiotic safety. Timing matters less than consistency, though our guide on when to take probiotics covers the nuances.

When to speak to a healthcare professional

Supplement labels cannot diagnose anything. Speak to your GP or pharmacist before starting a live-culture supplement if you are significantly immunocompromised, are receiving chemotherapy or immunosuppression, have a central venous catheter, have short bowel syndrome, or are seriously unwell.

Seek medical advice promptly — regardless of any supplement — if you have blood in your stool, unexplained weight loss, persistent change in bowel habit lasting more than three weeks, difficulty swallowing, persistent vomiting, or a new abdominal mass. Our guide on when to see a GP about stomach symptoms covers this in more detail, and gut health in the UK gives broader context.

Frequently asked questions

What is the difference between CFU and AFU?

CFU counts bacteria that can grow into a colony on an agar plate; AFU counts bacteria with an intact cell membrane using flow cytometry under ISO 19344. CFU measures the ability to reproduce, AFU measures structural viability. They are not interchangeable units.

Is AFU better than CFU?

AFU is faster and more reproducible, and it captures viable-but-non-culturable cells that plate counts miss. CFU proves reproductive capability and is the unit used in virtually all clinical trials. Neither is universally better — the most useful labels report both.

How do I convert AFU to CFU?

You cannot, reliably. There is no established conversion factor. The relationship depends on the strain, the manufacturing process and how far into shelf life the product is, and it typically widens over time.

Why is my probiotic's AFU number so much higher than other products' CFU?

Because flow cytometry counts clumped cells individually and includes cells that are alive but temporarily unable to grow on agar. The material may be identical; the yardstick is different.

Which probiotic brands use AFU?

AFU labelling is most common among newer brands and among products containing hard-to-culture organisms such as Akkermansia muciniphila or spore-forming Bacillus species. Most established UK and European brands still label in CFU.

Does a higher CFU count mean a better probiotic?

Not on its own. Strain identity, whether the dose matches the published research for that strain, and whether the count is guaranteed to the end of shelf life are all more informative than the headline number.

What does TFU mean on a probiotic label?

TFU stands for total fluorescent units — the sum of live and dead cells detected by flow cytometry. It is not a viability measure. A TFU figure will always be higher than the AFU figure for the same sample.

Why don't UK probiotic labels say "probiotic"?

UK government guidance treats "probiotic" as a health claim, and no such claim is currently authorised on the GB Nutrition and Health Claims Register. UK products therefore tend to say "live cultures" or simply list the strains.

What does "guaranteed to end of shelf life" mean?

It means the stated count is the minimum you should still receive on the expiry date, rather than the amount present on the day of manufacture. It is a substantially stronger commitment.

Should I choose a probiotic based on the unit alone?

No. Use the unit to make like-for-like comparisons, then judge the product on strain designation, per-strain dosing, potency guarantee, storage requirements and third-party testing. Our guide to choosing a probiotic walks through the full process.

References

1. International Organization for Standardization. ISO 19344:2015 — Milk and milk products: Quantification of lactic acid bacteria by flow cytometry.
2. Hill C, Guarner F, Reid G, et al. ISAPP consensus statement on the scope and appropriate use of the term probiotic. Nature Reviews Gastroenterology & Hepatology, 2014;11:506–514.
3. International Scientific Association for Probiotics and Prebiotics. Decoding a probiotic product label.
4. Wilkinson MG, et al. Insights into the enumeration of mixtures of probiotic bacteria by flow cytometry. BMC Microbiology, 2023.
5. Pane M, et al. Flow cytometry: evolution of microbiological methods for probiotics enumeration. Journal of Clinical Gastroenterology, 2018;52(Suppl 1):S41–S45.
6. Michalak M, et al. Dynamics of active fluorescent units and water activity changes in probiotic products. 2023.
7. Zheng J, Wittouck S, Salvetti E, et al. A taxonomic note on the genus Lactobacillus. Int J Syst Evol Microbiol, 2020;70:2782–2858.
8. Department of Health and Social Care. Nutrition and health claims: guidance to compliance with Regulation (EC) 1924/2006. GOV.UK.
9. Committee of Advertising Practice. Food: probiotic claims. ASA/CAP advice online.
10. NHS. Probiotics.
11. Elgorriaga-Islas E, et al. Assessing bacterial viability and label accuracy in human and poultry probiotics sold in the United Kingdom. Microorganisms, 2025;13(8):1933.
12. Sionek B, et al. Label accuracy of commercial probiotics — CFU enumeration and microbial composition. Annals of Microbiology, 2026.
13. Beloborodova E, et al. Specific cultivation-independent enumeration of viable cells in probiotic products using FISH and flow cytometry. 2024.
14. Vecchione A, et al. Analysis of the microbial content of probiotic products commercialised worldwide. Frontiers in Microbiology, 2023.
15. Eurofins. Flow cytometry for probiotic enumeration — Q&A with the experts.

About the author

Dr Zeeshan Afzal is a qualified medical doctor (MBBS) and Medical Officer at Welzo, where he reviews and authors clinical content across digestive health, nutrition and preventive medicine. He writes to help UK readers interpret health information and product labelling accurately, and reviews Welzo's gut health content against current guidance and peer-reviewed literature.

Medical disclaimer

This article is for general information and education only. It is not medical advice and does not replace consultation with a qualified healthcare professional. Food supplements are not medicines and are not intended to diagnose, treat, cure or prevent any disease. If you are pregnant or breastfeeding, taking prescription medication, immunocompromised, or living with a diagnosed medical condition, speak to your GP or pharmacist before starting any supplement. If you have concerning digestive symptoms, contact your GP or call NHS 111.