What 9 Years of Microbiome Data Now Lets Us Tell Clinicians

For most of the last decade, the honest answer to “will this probiotic work for my patient?” was a shrug. After nine years and more than 18,000 patients, it isn’t anymore. Today we’re sharing what our data now lets us say — and inviting clinicians and researchers to scrutinize it.

The question nobody could answer

Walk into any pharmacy and the probiotic shelf makes a hundred promises. Walk into a clinic and you’ll find most providers quietly skeptical — not because they doubt the microbiome matters, but because they couldn’t answer the only question that counts at the point of care: will this work for the person in front of me? Strain claims pulled from isolated trials in unrelated populations don’t answer that. So clinicians hedged, or didn’t recommend at all.

What we did differently

Starting in 2015, we took a different approach: sequence the patient, formulate against what their microbiome actually shows, and — the part almost nobody does — follow up and record what happened. We did that for nine years, across thousands of patients, pairing the before with the after. That pairing is the hard part, and it’s the foundation of the Central Intelligence Engine that powers Flore Clinical today.

What the data now says

• Nine years of paired records: 18,392 patients formulated, 14,704 sequencing subjects across 23,000 tests, and 651 paired before/after outcome records — across 200+ conditions.
• In real-world follow-up, many patients report symptom improvement across successive formulation cycles, and retesting can show measurable microbiome change over time.
• Because this is observational data with no control arm, Flore does not publish condition-level efficacy percentages — these patterns are hypothesis-generating, not proof of efficacy.
• Many patients present with symptoms across two or more body systems, which is why the data supports multi-formula and booster-layered recommendations rather than one blunt product.

What it means for clinicians

It means the answer to “will this work for my patient?” is no longer a shrug. When you submit a patient’s microbiome data, we can match it against the patients who looked like them and the formulations they were given. The ingredients are selected by the data, not by suppliers and not by a single paper. Three strains — Bifidobacterium breve, Lactiplantibacillus plantarum, and Lactobacillus rhamnosus — earned a place in every adult formula by appearing in at least 30% of successful formulations across every body system.

We’re putting this out for scrutiny

This is real-world, observational evidence, and we’re explicit about its limits: there is no control group, symptom resolution is self-reported, and the paired cohort is subject to retest selection bias. The right next step is independent review and controlled trials — and randomized controlled trials are in development for The Regular One (GI) and The Bright One (mood/neuro). We’re inviting MDs, GI and integrative clinicians, researchers, co-investigators, and IRB partners to read the methods, challenge the work, and collaborate.

Read the white paper · See the full evidence · Become a peer reviewer or research partner · Join the provider portal

These statements have not been evaluated by the Food and Drug Administration. GoodOnes™ products are not intended to diagnose, treat, cure, or prevent any disease. Real-world evidence is observational and uncontrolled.