How to Choose a Probiotic: CFU Count, Strains, and What Actually Matters

Higher CFU count does not mean a better probiotic. The evidence-supported range is 1-50 billion CFU for most applications. What matters more: specific strain identification (genus + species + strain number), clinical evidence for that strain at that dose, guaranteed viability through expiration (not just at manufacture), delivery mechanism (delayed-release capsules survive stomach acid better), and third-party testing. A 10 billion CFU product with the right strain for your condition will outperform a 100 billion CFU product with random, unstudied strains.

When buying a probiotic, the number of bacteria (CFU) on the label is not the most important thing. What matters more is which exact type of bacteria is in the product and whether scientists have actually tested that type for your specific problem. Look for the full name of the bacteria (it should have three parts, like a first, middle, and last name). Also check that the product promises the bacteria will still be alive when you take it, not just when it was made.

At a Glance

Property	Value
Evidence Level	Moderate (quality criteria well-established by ISAPP and clinical guidelines; specific product comparisons limited)
Primary Use	Rational probiotic supplement selection
Key Principle	Strain specificity and clinical evidence trump CFU count and marketing claims

How to Choose a Probiotic: Cutting Through the Marketing

The probiotic supplement aisle is one of the most confusing in the entire supplement market. Products range from 1 billion to 500 billion CFU. Some list 15 strains; others list 1. Prices range from $10 to $80 for a month’s supply. Marketing claims range from modest (“supports digestive health”) to absurd (“transforms your gut microbiome”).

Most of this is noise. Here is what actually matters when choosing a probiotic, ranked by importance.

The 5 Things That Matter (In Order)

1. Strain Identification: The Most Important Factor

A probiotic label should identify each organism with three levels of nomenclature:

Genus: Lactobacillus, Bifidobacterium, Saccharomyces, etc.
Species: rhamnosus, infantis, boulardii, etc.
Strain: GG, 35624, CNCM I-745, etc.

Example of a good label: Lactobacillus rhamnosus GG — 10 billion CFU Example of a useless label: “Proprietary probiotic blend — 50 billion CFU”

The strain designation is what connects the product to clinical evidence. If a product does not list strain numbers, you have no way to verify whether any research supports its use. This is the single most important criterion.

For which strains are supported by evidence for which conditions, see my best probiotic strains by condition guide.

2. Clinical Evidence for the Listed Strain(s) at the Listed Dose

Once you have identified the strain, ask: has this specific strain been studied in human clinical trials for my condition, at the dose listed on the label?

Common disconnect: Many products list strains that have been studied, but at different doses than what the product provides. Bifidobacterium infantis 35624 was studied at 1 x 10^8 CFU (100 million) — products providing 10 billion CFU of this strain are extrapolating beyond the evidence, not following it.

Check: Search PubMed for the full strain designation. If you find zero human trials, that strain has no evidence-based clinical application — regardless of what the marketing says.

3. Guaranteed Viability Through Expiration

This is a critical quality indicator that many consumers overlook. The CFU count on the label can mean two different things:

“At time of manufacture”: The product had X billion CFU when it was made. By the time you buy it, a significant proportion may be dead — especially if storage conditions were not optimal.
“Through expiration” or “guaranteed at expiration”: The product will still contain the listed CFU count at the end of its shelf life. This is the standard you should demand.

The International Scientific Association for Probiotics and Prebiotics (ISAPP) recommends that labels state CFU “at end of shelf life” and specifies storage conditions (1).

What this means practically: A product listing “50 billion CFU at manufacture” that you buy 6 months later may contain 20 billion CFU — or fewer. A product listing “20 billion CFU through expiration” is a more honest and reliable product.

4. Delivery Mechanism

Probiotics must survive stomach acid to reach the intestine alive. The pH of the stomach (1.5-3.5) kills the majority of unprotected bacteria. Delivery technology matters:

Delayed-release capsules (DR or enteric coating): These capsules resist dissolving in the acidic stomach and release their contents in the higher-pH small intestine. Studies show 2-10x better viability compared to standard capsules.

Acid-resistant strains: Some organisms — particularly Saccharomyces boulardii and spore-forming Bacillus species — are inherently acid-resistant and do not require special delivery technology.

Standard gelatin capsules: Dissolve in the stomach. A significant proportion of bacteria are killed by acid exposure. If the product uses standard capsules, the CFU count needs to be higher to compensate for gastric losses.

My preference: Delayed-release capsules for Lactobacillus and Bifidobacterium species. Standard capsules are fine for S. boulardii and Bacillus coagulans.

5. Third-Party Testing

The supplement industry has a quality control problem. Multiple independent analyses have found that probiotic products often contain:

Fewer viable organisms than the label claims
Different species than what is listed
Contaminants not disclosed on the label

Look for products with independent third-party testing: NSF International, USP, ConsumerLab, or the manufacturer’s own published stability data.

The CFU Myth: More Is Not Better

This is the biggest misconception in the probiotic market. Let me be direct: there is no evidence that 100 billion CFU is better than 10 billion CFU for any clinical indication.

What the Evidence Actually Shows

The clinically effective dose range varies by strain and condition:

Strain	Effective Dose	Condition
B. infantis 35624	1 x 10^8 CFU (100 million)	IBS
L. rhamnosus GG	10-20 billion CFU	Immune support, diarrhea prevention
S. boulardii	250-500 mg (~5-10 billion CFU)	Diarrhea, C. difficile
L. helveticus R0052 + B. longum R0175	3 billion CFU (combined)	Mood, anxiety
L. plantarum 299v	10 billion CFU	IBS, bloating

Notice the range: from 100 million to 20 billion. The effective dose is strain-specific and based on clinical trial data, not a “more is always more” principle.

Why More Can Actually Be Worse

Excessive gas and bloating. Very high CFU counts (100+ billion) can overwhelm the gut’s capacity to accommodate new organisms, causing temporary but significant GI distress.
Competitive exclusion among probiotic strains. In multi-strain products with very high CFU counts, the dominant strains may suppress the less-competitive strains — defeating the purpose of including them.
Cost without benefit. You are paying a premium for CFUs that have no additional clinical evidence.

The practical range: 1-50 billion CFU covers every evidence-based probiotic application. Products above 50 billion are marketing, not medicine.

Red Flags on Probiotic Labels

Avoid if you see:

“Proprietary blend” without individual strain CFU counts — you do not know how much of each strain you are getting
No strain designations (only genus and species)
“At time of manufacture” for viability claims
Unrealistic health claims (“heals leaky gut,” “cures IBS,” “boosts metabolism”)
No storage instructions (many probiotics require refrigeration)
No expiration date

Look for:

Full strain designation for each organism
Individual CFU count per strain
“Guaranteed through end of shelf life”
Delayed-release or enteric-coated capsules
Third-party testing certification
Clear storage instructions
Expiration date

How to Read a Probiotic Label: Example

Good label:

Supplement Facts
Serving Size: 1 delayed-release capsule

Lactobacillus rhamnosus GG        10 billion CFU
Bifidobacterium infantis 35624    1 billion CFU
Saccharomyces boulardii CNCM I-745  5 billion CFU

Guaranteed through expiration when stored as directed.
Store refrigerated (2-8°C).
Third-party tested by ConsumerLab.

This label tells you exactly what strains you are getting, at what doses, with viability guaranteed through expiration, with clear storage instructions and third-party verification. Every claim on this label can be matched to clinical evidence.

Bad label:

Proprietary Probiotic Blend       100 billion CFU
Lactobacillus acidophilus, L. plantarum, L. rhamnosus,
B. lactis, B. longum, L. casei, L. paracasei,
L. salivarius, L. bulgaricus, S. thermophilus,
B. breve, L. fermentum, L. gasseri, B. bifidum,
L. helveticus

This label lists 15 species but no strain numbers, no individual CFU counts per strain, a proprietary blend that hides the actual formulation, and a headline CFU that may not be achievable for all listed organisms. You cannot connect any component of this product to clinical evidence.

My Recommendation by Situation

”I just want something for general gut health”

L. rhamnosus GG: 10 billion CFU/day
Simple, well-studied, widely available
Add prebiotic fiber (25-35g/day from diet) — see probiotics vs. prebiotics for the full picture

”I have IBS and bloating”

B. infantis 35624: 1 billion CFU/day (Align brand)
Or L. plantarum 299v: 10 billion CFU/day
Give it 4-8 weeks before assessing

”I’m taking antibiotics”

S. boulardii CNCM I-745: 250 mg twice daily
Start with the antibiotic course, continue 2 weeks after
Take S. boulardii at any time (resistant to antibiotics)

“I want mood and stress support”

L. helveticus R0052 + B. longum R0175: 3 billion CFU/day (Cerebiome)
Allow 30 days for effect

The Bottom Line

Choosing a probiotic should be a rational, evidence-based decision — not a response to marketing. Identify your specific condition, find the strain with clinical evidence for that condition, verify the strain and dose on the product label, and confirm viability through expiration. A 10 billion CFU product with the right studied strain will outperform a 100 billion CFU product with unstudied strains every time. This is what the research actually says, and it is what I tell my patients.

References

Hill C, Guarner F, Reid G, et al. Expert consensus document: the International Scientific Association for Probiotics and Prebiotics consensus statement on the scope and appropriate use of the term probiotic. Nature Reviews Gastroenterology & Hepatology. 2014;11(8):506-514. doi:10.1038/nrgastro.2014.66
Jackson SA, Schoeni JL, Vegge C, et al. Improving end-user trust in the quality of commercial probiotic products. Frontiers in Microbiology. 2019;10:739. doi:10.3389/fmicb.2019.00739
Gruenwald J, Graubaum HJ, Harde A. Effect of a probiotic multivitamin compound on stress and exhaustion. Advances in Therapy. 2002;19(3):141-150. doi:10.1007/BF02850271