A shelf-life statement looks like a single number printed on a bottle. Behind it sits a data package: a formula, a package, a storage condition, and a set of test results taken at known intervals. When a buyer asks "can you give me 24 months?", the useful reply is not yes or no — it is a description of the evidence that will support the number.

Three inputs decide the answer

Shelf life is a property of the product as packaged, not of the formula alone. It depends on three things at once: how stable the actives are in that matrix, how much moisture and oxygen the package lets through, and the temperature and humidity the product will actually meet in the market. Change any one of them and the number changes. The same gummy in a jar with an induction seal and in a stick pack can carry different date codes.

Real time is the evidence, accelerated is the screen

Accelerated studies expose the product to a fixed elevated condition — the ICH Q1A(R2) convention of 40 °C at 75 % relative humidity is the usual reference — and are used to rank formulations and catch failures early. They are a screening tool, not a substitute for the record. A commercial date code is normally supported by real-time data at the labelled storage condition, because acceleration can create degradation pathways that do not occur on the shelf, and can miss ones that only appear slowly.

This is why a supplier who has never made the formula before cannot honestly promise a long date code on day one. The honest sequence is: make the batch, start the clock, and grant the shelf life the data supports at the time of shipment.

What actually degrades, form by form

  • Gummies — moisture migration between the gummy and the headspace, sugar crystallisation, and loss of chew texture. Water activity is the single most predictive variable here; a gummy that starts too high has nowhere to go.
  • Softgels — oxidation of the fill oil, measured by peroxide and anisidine value, and cross-linking of the shell. Light and trace metals accelerate both.
  • Hard capsules — shell brittleness as moisture drops, and hygroscopic actives pulling water through the shell.
  • Powders — caking and flow loss, plus vitamin potency drift in blends with reducing sugars.
  • Liquids — pH drift, colour change, and loss of preservative efficacy over time; a liquid that passes at t=0 can fail at month six.
  • Probiotics — viable count, which is dominated by water activity and by the package's moisture barrier rather than by the strain alone.

The package is part of the formula

HDPE, PET and glass differ in oxygen and moisture transmission; a foil stick pack is a different proposition from a jar; a desiccant only works if the closure keeps the headspace dry. When a project moves from a sample bottle to production packaging, the stability record has to move with it. A common and expensive mistake is to run stability in one container and ship in another.

Overages: legitimate, but they must be declared correctly

Many formulas carry a deliberate overage so that the label claim still holds at the end of shelf life rather than only on the day of manufacture. That is normal practice, sized to the measured degradation curve. What is not acceptable is an overage used to cover a weak assay at t=0, or a label that declares a number the product never contained.

A testing plan you can ask for

A workable design is: potency, appearance and water activity at 0, 3, 6, 9, 12, 18 and 24 months on real-time storage; microbiological testing at the start and the end; and accelerated points used to catch problems early rather than to set the date. Ask for the condition, the package, the number of batches, the test methods and who performed them. A single batch is a data point, not a shelf life.

Five questions to ask before you print a date code

  1. Which storage condition does the claim refer to, and does it match my market's climate?
  2. Was the study run in the production package, including the closure and any desiccant?
  3. How many batches, and were they at the commercial scale?
  4. Which markers were measured — assay only, or also moisture, colour and micro?
  5. What happens at the end of the claimed period: does the label claim still hold?

How Haiyue Shengda handles it

We set the date code from the formula and package we are actually shipping, run real-time points alongside accelerated screening, and issue the stability summary with the batch documents. Trial orders start at 100 units, samples ship in 3–5 days, production follows within 7 days, and every batch is accompanied by a COA. See our quality and compliance page, or contact us with your target market and pack format — we reply within one hour.