The Date on Your Pack Is a Promise, Not a Guess: Why Kenyan Food Factories Should Stop Treating Shelf-Life as a Launch Decision

By Isaac Muthui, CEO, Tamalaki Business Network Ltd. Originally published at factoraerp.com.

Every pack that leaves a food factory carries a promise printed in a few characters: "Use by", "Best before", a day and a month. Customers, supermarkets and inspectors read that date as a statement of fact. Inside many factories, it is closer to a guess. The number was set when the product was launched, perhaps borrowed from a similar product or a supplier's suggestion, and nobody has tested it since the recipe, the packaging or the distributor changed.

Shelf-life validation is the work of proving that date. It means describing the product, understanding what limits its life, testing it under the conditions it will actually meet, choosing a date with a margin of safety, and keeping the evidence so you can show anyone how you got there.

This shelf-life validation checklist for food factories is written for dairies, juice and beverage plants, bakeries, meat and ready-to-eat food processors, snack makers and other Kenyan and East African food and agro factories. It does not replace your food technologist, microbiologist, laboratory or certification body. It sets out ten practical checks for turning the date on your pack into something you can defend.

What the guidance actually says

Kenya's food hygiene regulations. The Food, Drugs and Chemical Substances (Food Hygiene) Regulations, made under Cap. 254, require that each container "be embossed or otherwise permanently marked in code or in clear to identify the producing factory and the lot", and that specific products, as specified by the Minister, "bear prominently a date-marking, showing the last day, month and year" the product may be sold (regulation 14).

Codex on labelling. The Codex General Standard for the Labelling of Prepackaged Foods (CODEX STAN 1-1985) sets out how a date should be declared: at least the day and the month for products with a minimum durability of not more than three months, and the month and the year for products with a minimum durability of more than three months (4.7.1). It adds that "any special conditions for the storage of the food shall be declared on the label if the validity of the date depends thereon" (4.7.2).

Codex on hygiene. The Codex General Principles of Food Hygiene, CXC 1-1969, lists "the intended shelf-life of the product" among the things time and temperature control systems should take into account, alongside the food's water activity and pH, packaging and intended use. It also says records "should be retained for a period that exceeds the shelf-life of the product or as determined by the competent authority" (13.4).

Codex on Listeria. The Codex Guidelines on the Control of Listeria monocytogenes in Foods, CXG 61-2007, says that because the organism can grow at refrigeration temperatures, "the length of the shelf-life should be based on appropriate studies that assess the growth of L. monocytogenes in the food", and that such studies "should account for the fact that appropriate low temperatures may not be maintained throughout the entire food chain until the point of consumption".

A detailed public benchmark. The Food Safety Authority of Ireland's Guidance Note 18, Validation of Product Shelf-life (Revision 5) is not Kenyan law, but it is one of the clearest step-by-step public guides. It defines shelf-life as "the period of time over which a food maintains its safety and/or quality under reasonably foreseeable conditions of distribution, storage and use", and says that if a shelf-life is too long, or assumes unrealistic conditions, "there is an increased risk of food safety issues arising, people becoming ill and damage to the food business operator's brand and reputation".

Why shelf lives go wrong

Shelf lives rarely fail because a factory has no date on the pack. They fail because the date was never tied to evidence, or the evidence stopped matching the product.

The patterns are familiar. The shelf-life was copied from a competitor's pack or a similar product. A study was done once, at a perfect 4 Â°C in the lab fridge, on samples from the best batch of the month. The recipe was later changed to cut sugar or salt, or a cheaper film was approved, and nobody asked whether the date still held. The distributor's van, the supermarket chiller and the customer's fridge run warmer than the study assumed. And when a complaint about a sour or swollen pack arrives a week before the date, nobody connects it to the shelf-life decision.

The 10-step shelf-life validation checklist for food factories

1. Decide what the date protects: safety or quality

Start by deciding which kind of date the product needs. FSAI Guidance Note 18 explains that a "use by" date applies to foods that are likely, after a short period, to pose an immediate risk to health from a microbiological point of view, such as chilled ready-to-eat foods, raw meat and fish and fresh juices. A "best before" date applies where quality, not safety, is the issue, such as many canned, dried, ambient and frozen foods.
For a use-by product, the study is about pathogens. For a best-before product, it is mainly about spoilage, texture, flavour, rancidity and appearance. Write down which one you are proving.

2. Describe the product and what limits its life

Write a short product description: recipe, process, kill step if any, packaging, storage temperature, intended consumer and how it will be used.
List the characteristics that control its life. FSAI groups them as intrinsic (pH, water activity, antimicrobial substances, microflora, formulation) and extrinsic (temperature, packaging, gas atmosphere, humidity, hygiene, distribution and consumer practices), and notes that for most businesses the most important are pH, water activity, storage temperature and packaging.
Record the target values with acceptable ranges. If the pH or water activity drifts outside them, the shelf-life evidence no longer applies. Your HACCP CCP monitoring records are where those values should be checked.

3. Set the shelf-life before launch, and know when to redo it

FSAI says shelf-life should be estimated during product development and set before the food goes on sale, at the point where you are confident you can produce the same food consistently from batch to batch under real processing conditions.
Redo the work when there is no evidence for an existing product's date, and whenever the food or its production is modified or reformulated. Keep a list of the changes that trigger a review: recipe, supplier of a key ingredient, packaging material, process temperatures or times, and storage or distribution route.

4. Test the worst case, not the best batch

FSAI recommends that study samples represent the worst-case formulation (for example, in an acidic product, the pH closest to neutral) processed under worst-case conditions (for example, the longest holding time before final processing).
Take samples from normal production runs, packed in the commercial packaging, not from a pilot batch made by the most careful person in the plant.
In the absence of a specified sampling plan, FSAI recommends at least three replicate samples from three separate production batches at each test interval.

5. Store samples the way the product will really be stored

FSAI says studies should mirror "reasonably foreseeable conditions of storage, distribution and use", using temperatures that reflect normal practice even where these differ from the recommended temperature, and that businesses must consider temperature abuse and justify the temperatures used.
CXG 61 makes the same point for chilled ready-to-eat foods: low temperatures may not be maintained all the way to the consumer.
In practice, that means using what you know about your own chain: the dispatch bay, the van, the distributor's store, the retailer's chiller and the customer's fridge. Your cold chain temperature monitoring records are the best evidence you have.

6. Run the study past the target date, at sensible intervals

FSAI suggests running a durability study up to and beyond the target shelf-life. Its example: for a 28-day target, test for 28 days plus seven.
Set test intervals by how perishable the product is. FSAI gives the example of daily testing for a seven-day product, and twice-weekly or weekly testing for less perishable foods.
Agree in advance what you will test at each interval (for example, total counts, spoilage organisms, relevant pathogens, pH, water activity, sensory checks) and what result ends the shelf-life.

7. Use the right kind of study for the hazard

A durability study tests the product as made, with whatever organisms are naturally present. FSAI notes that this has a limit: the pathogen of concern may not be present in the batches you test.
For ready-to-eat foods that can support Listeria growth, FSAI says a durability study on its own is not suitable to validate the shelf-life, and recommends combining it with challenge tests or predictive microbiology. A challenge test deliberately adds the organism to see whether and how fast it grows. It needs a competent laboratory.
CXG 61 notes that in some products single factors, such as a pH below 4.4, a water activity below 0.92, or freezing, may be relied on to prevent Listeria growth, and that combinations of factors should be validated.
FSAI also warns that predictive models are a guide and should be verified by a durability study or challenge test.

8. Read the results carefully and apply a margin of safety

Have results interpreted by a trained food microbiologist together with the testing laboratory, as FSAI recommends, and compare them first with legal limits, then with relevant standards and guidelines.
Don't rely on the average result. FSAI recommends using a higher percentile, such as the 90th or 95th, which is only possible with at least three replicates per time point.
Apply a margin of safety. FSAI says shelf-life "will never be an absolute value that terminates at an exact time and date", strongly recommends a margin of safety, and asks businesses to document the reasoning behind it.

9. Put the date and storage conditions on the pack correctly

Print the date in the format Codex describes: day and month for products that last up to three months, month and year for longer-life products, introduced by the right words.
If the date depends on storage, say so on the label, as CXS 1 requires. "Keep refrigerated at or below the stated temperature" is part of the shelf-life, not decoration.
Mark every pack with the factory and lot, as Kenya's food hygiene regulations require. Our label control checklist covers how to stop the wrong date or lot being printed, and our line clearance checklist covers changeovers between products with different dates.

10. Keep a shelf-life file and keep verifying

FSAI strongly recommends documenting all work on estimating, setting and validating shelf-life, filed together as part of your HACCP-based procedures, so that customers and inspectors can see the evidence. Keep it with your other food safety records, and keep production records for longer than the product's shelf-life, as CXC 1 recommends.
Verify periodically. FSAI says it is good practice to verify safety and shelf-life periodically, because subtle changes can arise in products over time, and to test samples taken from the distribution chain or following customer complaints.
Link the date to how you move stock. Dispatch first-expired, first-out, as in our FEFO and expiry tracking guide, and treat complaints about spoilage before the date as shelf-life evidence. Our customer complaint handling checklist shows how to route them.

A simple shelf-life file

Section	What to record
Product description	Recipe version, process, packaging, storage temperature, intended use
Date type	Use-by (safety) or best-before (quality), and why
Controlling factors	Target pH, water activity, salt, sugar, preservatives, with acceptable ranges
Study design	Batches and replicates, storage temperatures and why, intervals, tests, end-point criteria
Laboratory	Who tested what, by which method
Results	Results at each interval, by batch and replicate
Decision	Shelf-life set, margin of safety applied, and the reasoning
Label	Date format, storage instruction, lot coding
Review triggers	Recipe, supplier, packaging, process or distribution changes; complaints
Verification	Dates and results of periodic checks

If you cannot explain the decision row in two sentences, the shelf-life is not ready.

An illustration

This is an illustrative example, not a real case. A small juice processor prints a 21-day use-by date on a chilled juice, a figure set at launch. When it moves to a lighter bottle and a new distributor, the quality manager opens the shelf-life file and finds only a single page with the original figure. The team runs a durability study on three normal production batches, three replicate bottles each, stored partly at the cold-room temperature and partly at the warmer temperatures recorded in the distributor's van and at retail. The samples stay within the agreed limits for some time but not for the full 21 days under the warmer conditions. With the laboratory's advice, the team sets a shorter date with a margin of safety, adds a clear storage instruction to the label and schedules a verification test from retail samples. The shorter date is not a loss. It is the first date the factory can defend.

How FactoraERP helps

In FactoraERP, stock items can be set to track expiry with a shelf-life in days, and finished-goods lots are given an expiry date calculated from their production date. Dispatching from a later-expiring lot while an earlier one is still sellable needs an operations manager or admin to give a written reason, and the system records who did it, when and why. The laboratory module records test results against specification limits with a pass or fail status, so shelf-life checks sit in the same system as the batches they describe. See factoraerp.com.

Free download: eTIMS readiness checklist for factories

Kenyan manufacturers increasingly need their operational records, stock records and invoices to tell one consistent story. If you want a free eTIMS readiness checklist for factories, sign up for the FactoraERP Insights newsletter at https://isaacmuthui.substack.com. You will get practical operations notes for food and agro manufacturers. No spam; unsubscribe anytime. For background, see our guide to eTIMS for food and agro manufacturers.

FAQ

What is shelf-life validation?
Obtaining and documenting evidence that a food will stay safe and of acceptable quality until the end of the shelf-life on its label, under reasonably foreseeable conditions of distribution, storage and use. FSAI Guidance Note 18 describes it as proving that the shelf-life of a food is accurate.

What is the difference between a use-by date and a best-before date?
A use-by date marks the end of the period in which a perishable food can be safely eaten, and is used where the food could become a health risk after a short time. A best-before date marks the end of the period in which a food keeps its expected quality when properly stored; it may still be safe after that date.

Does Kenyan law require date marking on food?
Kenya's food hygiene regulations require every container to be marked to identify the producing factory and the lot, and require specified products to carry a date marking. Product standards and labelling rules may add more. Check your obligations with KEBS, your county public health office and your certification body.

How do you determine the shelf-life of a food product?
Describe the product and the factors that control its life, then test samples from normal production batches stored under realistic conditions, past the target date, at set intervals. For ready-to-eat foods that can support Listeria growth, combine this with challenge tests or predictive models. Set the date with a margin of safety and record the reasoning.

How many samples are needed for a shelf-life study?
It depends on the product and the sampling plan. Where no plan is specified, FSAI Guidance Note 18 recommends at least three replicate samples from three separate production batches at each test interval.

When should a shelf-life be re-validated?
When the recipe, a key ingredient, the packaging, the process or the distribution route changes, when there is no evidence behind an existing date, and when complaints or verification tests suggest the product is not lasting as long as the label says. Verify periodically even when nothing has changed.

Isaac Muthui, CEO, Tamalaki Business Network Ltd