Pet Supplement Stability Testing: How to Plan Shelf-Life Verification

Define which product the shelf-life decision covers

Shelf life belongs to a particular product configuration, not to a formula name. “The same probiotic powder” is not a complete description if the formula, manufacturing process, container, seal or fill quantity has changed.

Start with a versioned product description that identifies:

• The intended species, product format and target market.
• The formula and manufacturing-process versions, including the ingredients or properties covered by label claims.
• The commercial packaging configuration: container or film, closure, seal, fill quantity and relevant headspace (the unfilled space inside the pack).
• The proposed storage conditions, distribution route and normal use after opening.
• The shelf-life target to investigate and the evidence already available.

Keep a proposed shelf life separate from a conclusion supported by results. A target helps define the study; it does not become verified because it appears in a brief or supplier specification.

Early samples still have a useful role. They can reveal ingredient interactions, physical changes or unsuitable packaging before you commit to a larger study. For a shelf-life conclusion, establish how the tested samples represent the commercial formula, process and pack. Record any differences and why the data remain relevant.

In one cooperating manufacturer’s project, a botanical powder passed early sampling without issue; the production batch then moved from foil-lined bags to a plain laminate film and developed caking within two months on the shelf. The formula had not changed — the configuration had. That is the gap a versioned product description exists to catch.

If you are still deciding what the sample approval covers, use our pet supplement sample review guide. Keep those approval decisions connected to the stability plan as the product develops.

A batch certificate of analysis, or COA, provides results for the identified sample and tests. It can contribute to an initial baseline; data collected over time are needed to assess change during storage. Our supplement documentation guide explains the different roles of COAs, specifications and test reports.

Choose stability studies by the question they answer

The study combination should follow the decisions you need to make. These four approaches address different parts of the product’s life:

Accelerated data are useful for early decisions when the prediction is appropriate for the formulation. Elevated stress can also produce changes, such as melting or phase separation, that do not represent normal storage. Treat the model and its assumptions as part of the evidence, and verify projections with long-term data. There is no fixed time-conversion rule that establishes shelf life for every pet supplement. The IADSA stability guide, particularly sections 8–10, explains these study-design distinctions. As a human-supplement study-design reference, IADSA recommends at least two batches for an in-use study, with one chosen towards the end of shelf life, and examining the configuration in which significant changes are most likely — such as the largest pack size in a range (section 8.4).

Define “normal storage” precisely enough to study it. The plan should reflect the intended markets and supply route, with temperature and humidity recorded during storage. Opening a pack, shipping it through a hot route and keeping it sealed in a warehouse are different questions; specify which ones matter to this project.

One manufacturer has seen soft chews sweat oil and separate at 40 °C accelerated conditions while parallel real-time samples at intended storage remained intact through the same period. The accelerated failure was real, but the real-time samples did not show the same change during that observation period. If elevated stress changes the failure mechanism, a projection model based on that mechanism may be unsuitable; check its applicability first.

Select stability measures for the formulation and format

Select attributes that can detect the changes that would limit the product’s useful life. A good-looking product may have lost a claimed ingredient; a product that retains its measured content may still have an unacceptable texture or package failure.

Use the following table to select topics for technical review. The rows are candidate areas, not a requirement to run every test on every product.

A probiotic powder may need both the powder and live-microbe considerations. Formulation risks overlap; choosing one format row should not exclude an ingredient-specific issue.

Moisture content and water activity answer different questions. Moisture content measures the total water present. Water activity concerns the availability of water and helps assess conditions for microbial growth and other changes. Two products can have similar moisture content and different water activities. Use the measurements appropriate to the product rather than substituting one for the other. Manitoba Agriculture explains this distinction; FEDIAF’s pet food safety guide, Annex 2, gives pet food examples connecting water activity, storage and packaging integrity.

For numerical context on microbial growth conditions, Manitoba’s table places some low-water-activity-tolerant moulds and yeasts at about 0.60–0.62. About 0.6 is not a product stability acceptance limit: restricting growth does not establish that organisms have been killed, or that ingredient, texture or oxidation measures pass.

In a reptile powder project Anivor coordinated, the cooperating manufacturer's production batches developed fine-particle segregation: the lighter, low-density ingredients gradually migrated to the top of the pack and the powder layered, so each scoop delivered an uneven calcium dose.

Over the same period, as the weather turned into the humid rainy season, the project's recorded water activity climbed from about 0.2 to about 0.6, alongside observed caking and a 60% fall in the product's feed-intake rate. These were concurrent project observations, not a single-factor causal trial. The problem was resolved through the manufacturer's encapsulation and anti-caking process combination. It is also a reminder that a visually normal powder still needs segregation and per-serving content uniformity on its stability attribute list.

For a live-microbe product, identify when the stated count applies. A count at manufacture does not describe the count at the end of storage. Record the organism identity, reporting basis, method and required time points. For counts reported as colony-forming units (CFU), state the basis, such as CFU per gram or per serving. For mixtures, establish what the method can distinguish. The EFSA feed-additive guidance, section 2.4.1, describes stability tracking using active substances or agents; ISAPP’s label explanation discusses the importance of viable numbers through shelf life.

For probiotic viability, the IADSA human-supplement guide (Annex IV) strongly recommends real-time studies because accelerated survival data are difficult to extrapolate.

For one probiotic project, the cooperating manufacturer set the production count at three times the label claim with the aim of keeping the end-of-shelf-life count at or above it. Whether that overage is right for a formula is a study question — but the team could only frame the question because the specification stated which time point the count had to serve.

For every selected measurement, check that the method works in the finished-product matrix. A method suitable for a raw ingredient may face interference in a blend, chew or liquid. Agree the sample preparation and reporting basis so that results from different checkpoints can be interpreted together.

Put the plan in writing before testing starts

Copy the planning table below into your project document. Complete it with the relevant manufacturer, laboratory and technical reviewer. Assign those roles explicitly rather than assuming that the party supplying a sample also approves its shelf life.

Write criteria that someone else can apply consistently. “Looks acceptable” is too vague unless it refers to defined observations and a reference. For a powder, identify the agreed handling or dispersibility procedure and its acceptable outcome. For a hypothetical powder intended for dispersion in water, an internal project example could read: 1 g of powder into 50 mL water at 25 °C, stirred for 30 seconds, then passed through a 20-mesh sieve with no visible clumps. This hypothetical entry illustrates procedural specificity; it is not a validated method. For a powder used only by mixing into feed, define handling and limits around that actual use. For a live-microbe specification, identify the required count, units and time point, together with the method used to assess it. Put the actual values and their basis into the project plan.

Similarly, a sampling schedule needs a reason. Connect its checkpoints and sample quantities to the proposed life, expected changes, analytical needs and any follow-up or repeat work. The CFIA shelf-life study guidance, especially its planning and evaluation steps, is a useful methodological reference for connecting these decisions.

The completed plan should let a person outside the launch discussion identify exactly what is being tested, how it will be judged and which decision the results can support.

Example: a probiotic powder moves from a bottle to a pouch

Hypothetical planning example: a dog probiotic powder keeps the same formula, while the brand proposes changing from a bottle to a resealable pouch. Assume supporting bottle data exist, but corresponding pouch data have not yet been evaluated.

An illustrative checkpoint schedule for this change: T0, 3, 6, 9, 12, 18 and 24 months for the real-time study. Set separate accelerated checkpoints for the stress conditions and purpose, and in-use checkpoints for the actual use period; compare those results with relevant real-time data. Adjust the rhythm to the proposed life and the changes you expect.

For this pack-change example, check two key pieces of evidence first: the barrier performance of the new film and the viable count after repeated opening. Assess them alongside the existing ingredient data.

1. Describe the change. Keep the formula and process versions visible, and identify the new film, seal, reclosure, fill quantity and expected use. A matching ingredient list does not make the packaging configurations identical.

2. Compare the exposure assumptions. Review the relevant moisture and oxygen barriers, seal performance, headspace and repeated-opening conditions. Packaging specifications help describe the change; the review must also establish what those differences mean for the product.

3. Check what the existing evidence covers. Map the bottle study’s product version, conditions, viable-count results and relevant powder/pack observations against the proposed pouch. Identify which parts remain applicable and where additional evidence is needed. Have the technical reviewer determine the justified scope of supplementary or new studies.

4. Record the decision before carrying over the date. If the evidence is judged applicable, record which parts of the bottle study transfer to the pouch and under what conditions. Where it does not support the new configuration, keep the shelf-life statement proposed while the required evidence is obtained and reviewed. Record the approved configuration and any storage or use changes when the decision is made.

This process produces a documented decision about the new pack, rather than an assumption based only on an unchanged formula. For the separate design handoff, use our pet supplement packaging brief.

Use the results to make a shelf-life decision

In Anivor’s coordination work, the consequences of skipping this step can also appear in a customer’s warehouse, a platform review or a retailer audit — as caked powder, a viable count below the label claim, or a request for data the file does not contain. Rebuilding evidence afterwards can add cost and waiting time: some studies must restart with new samples, while existing evidence that remains applicable can be retained after review.

Start by checking whether the study samples meet the relevant initial specifications and whether the methods and sample identities match the plan. Keep that baseline connected to later measurements.

Review the attributes individually and together. The proposed shelf life is limited by whichever relevant attribute first fails its acceptance criterion; passing results elsewhere do not cancel that failure. In one cooperating project, other attributes passed, but a slow water-activity climb in the largest pack size limited support for the 24-month target; after review, the team set the shelf life at 18 months. The decision follows the relevant attributes and their acceptance criteria rather than averaging the passing results elsewhere; a rise in water activity alone does not automatically mean failure. Investigate unusual results and storage deviations, document their effect on interpretation, and preserve the reasoning behind any action.

The resulting decision may be to:

• Support a shelf life for the defined configuration and conditions, with any continuing work specified.
• Investigate a failure or trend, revise the formula, process or pack, and define the further evidence needed.
• Keep the decision open because the data, methods or study coverage do not yet support the proposed statement.

The outcome should identify the product and pack versions, conditions, evidence reviewed, limitations that affect use, and the person responsible for the technical decision. Retain the plan, data, deviations and conclusion together so that a later change can be assessed against the same record.

Include the study scope in your first-order project budget.

Frequently asked questions