Live organisms decline. That is not a defect; it is the nature of the material. Which means a colony-forming unit count is only meaningful when paired with the moment it was measured — and the two moments the industry uses are very far apart.
At time of fill
A count taken at fill tells you what went into the capsule or sachet on the day it was made. It is a legitimate manufacturing control and a genuinely useful number for a production team. It is also the most flattering number the product will ever produce, because it is measured before any of the conditions that reduce viability have had time to act.
At expiry
A count supported through to expiry is a claim about what the consumer actually swallows on the last day the label says the product is good. It requires knowing how the specific strains in the specific matrix, in the specific packaging, behave over time — which cannot be inferred from a fill-day count.
This is the number most retailers and most informed consumers assume they are being told. When the two are quietly swapped, the claim is not technically false, but it is not the claim the reader thinks they are reading.
Why the strain matters, not just the count
Enumeration that does not distinguish strains produces a total. A total is not much use in a multi-strain blend, because the strains do not decline at the same rate. A product can hold its headline number while its clinically interesting strain quietly disappears behind more robust organisms in the same count.
Strain-specific enumeration — plating on selective media such as MRS agar and resolving organisms individually — is what makes a blend claim checkable rather than merely arithmetic.
Modeling the decline instead of guessing it
Supporting an expiry claim means measuring viability at intervals across a stability program and modeling the trajectory, rather than testing once and adding a margin. curí quotes stability CFU modeling as a separate stability program, across a horizon set by your shelf life, for exactly this reason; release certificates report the count at time of fill: the shape of the curve, not a single later data point, is what tells you whether a claim holds at month eighteen.
Two practical consequences follow. First, overage — the deliberate overfill that lets a product still meet its label at expiry — should be set from data, not convention. Second, packaging and moisture control are part of the claim, because they are part of the decline.
What to put on the label
Say which moment the number refers to, and be able to show the work behind it. “50 billion CFU at expiry” is a stronger claim than “50 billion CFU” precisely because it is more specific and more difficult — and a buyer who reads carefully will know that.
If the data only supports a fill-day number today, that is a reason to run the stability work, not a reason to leave the moment unstated.
This note describes curí’s own laboratory practice and is written for brand and regulatory teams. It is not regulatory advice on a specific product. All programs are designed to support structure/function claims permissible under DSHEA; curí does not design studies to support disease claims.