Shelf-life and accelerated stability testing for a Sri Lankan SKU
By the Silk Foods Ceylon Editorial Team
Buyer’s snapshot
- The Food (Labelling and Advertising) Regulations 2026, in operation from 1 July 2026, require a numerical date of expiry and make selling past it an offence.
- They say nothing about how that date is derived. The method sits with the manufacturer, the liability sits with the brand on the label.
- A 12-month claim takes 12 months to prove. Accelerated storage at 40 degrees Celsius and 75 percent relative humidity runs 6 months, with pulls at 0, 3, and 6.
- Silk Foods Ceylon (SFC) scopes the study inside R&D, which adds 4 to 6 weeks at the front, with SLSI clearance typically 4 to 8 weeks after that.
Since 1 July 2026 a packaged food sold in Sri Lanka has to carry a numerical date of expiry, and selling past that date is an offence under the Food (Labelling and Advertising) Regulations 2026. The regulations are precise about how the date is printed. They are silent about how it is decided.
That silence is where first-run projects get into trouble. The number on the pack is a claim the brand owner carries, and the only thing defending it is a study run on the product that actually goes into the jar. Few briefs arrive with that study. Several arrive with the labels already printed.
A date of expiry is a legal position, not a design choice
The Food (Labelling and Advertising) Regulations 2026 were gazetted on 26 June 2026 under the Food Act, No. 26 of 1980, and came into operation on 1 July 2026. They require the date of expiry to be declared in numerical form, preceded by Expiry, Use before, Use by or Best Before, printed at not less than 1.5 millimetres in height. Regulation 5(4) then bars the sale, offer, or transport for sale of any food after that date.
Two details in the same gazette catch local brand owners out. Where only a month and a year are given, the date of expiry is deemed to be the last day of that month, so a January 2027 marking expires on 31 January and not on some later date the sales team assumed. And food supplied business to business is exempt from most labelling requirements but not from ten mandatory items, which include the date of manufacture, the date of expiry, and the batch number, with the words Not for retail sale printed in bold at not less than 3 millimetres.
A bulk pack moving between two manufacturers therefore still needs a defensible expiry date, and a distributor buying 25 kilo sacks for repacking inherits it rather than resetting it.
The clearance path that sits alongside this is set out in SLSI step-by-step for a first packaged-food submission in Sri Lanka.
Why does a full-term study take as long as the claim?
Because a shelf life is an observation, not a calculation. The Codex Alimentarius Commission defines the date of minimum durability, the best-before date, as the end of the period under stated storage conditions during which the product remains fully marketable and retains the specific qualities claimed for it. A use-by date carries a harder meaning: after it, the food should not be regarded as marketable. Proving either one to 12 months means holding samples for 12 months.
The Food and Drink Federation guidance, published in 2017 with a foreword by the then chairman of the UK Food Standards Agency, rules out the two usual shortcuts. Do not copy a shelf life from a similar product. Do not rely solely on prediction software. Run the trial on the same ingredients, equipment, and procedures as standard production.
That last rule is the expensive one. A recipe validated in a kitchen at 4 litres is not evidence for the same recipe at 1,500 jars through a retort: fill temperature, headspace, holding time, and cooling curve all move, and each moves the degradation clock. Pilot-sample data has to be confirmed on the first commercial batch.
The scale-up gap between a tested recipe and a commercial batch is worked through in reformulating a kitchen recipe for retort and 1,500-jar consistency.
What accelerated testing buys, and what it does not
Accelerated storage raises temperature to speed up the chemistry, then reads the result back onto a normal calendar. The Institute of Food Technologists described the underlying convention in September 2023: the Q10 value measures how much faster a chemical or biological system changes when the temperature rises by 10 degrees Celsius, so a system with a Q10 of two changes twice as fast at 30 degrees as it does at 20. Testing that would take months or years can be condensed to months or weeks.
The conditions themselves are borrowed from pharmaceutical practice. ICH guideline Q1A(R2) sets accelerated storage at 40 degrees Celsius plus or minus 2 and 75 percent relative humidity plus or minus 5 for a minimum of 6 months, with at least three time points, commonly 0, 3, and 6. Food manufacturers treat that as a working convention rather than a legal requirement, since no Sri Lankan regulation prescribes a protocol at all.
| Study type | Storage condition | Minimum period | What it settles |
| Long term | 25 degrees Celsius and 60 percent RH, or 30 degrees and 65 percent RH | 12 months | The claim itself. This is the evidence a 12-month date rests on |
| Long term, hot and humid market | 30 degrees Celsius and 75 percent RH | 12 months | The same claim under conditions a tropical distribution chain actually delivers |
| Intermediate | 30 degrees Celsius and 65 percent RH | 6 months | A middle reading when the accelerated result shows significant change |
| Accelerated | 40 degrees Celsius and 75 percent RH | 6 months, pulls at 0, 3, and 6 | An early read on failure mode and a provisional date, pending the full-term result |
Conditions in the table are the ICH Q1A(R2) general case, with the hot and humid long-term row taken from the World Health Organization stability annex published in 2018, which allows 30 degrees Celsius and 75 percent relative humidity for markets in that climatic band. The same annex warns that applying conditions suited to cooler zones to a product sold in a hot and very humid one risks a substandard product.
That warning is the local point. A dossier written for a European shelf assumes 25 degrees and 60 percent humidity. A Sri Lankan delivery van, a rural outlet without air conditioning, and a warehouse in the intermonsoon period do not.
The limits deserve as plain a statement as the benefits. Acceleration works on reaction rates, so it reads oxidation, browning, and vitamin loss reasonably well and reads physical failure poorly, because emulsion separation, staling, crystallisation, and moisture migration through a seal do not share one temperature relationship. The same 2023 article flags the shortcut directly: add 10 degrees and halve the time oversimplifies, since mechanisms in one product respond differently. Accelerated data narrows the question without closing it.
The packaging side of that problem, and why humidity sets the barrier spec here, is covered in herbal tea shelf life in Sri Lanka: moisture barrier specs.
What gets measured at each pull point
A pull point is a scheduled sample withdrawal, and each one costs money, so the parameter list should be short and drawn from the ways this particular product fails. A high-acid retort spread fails differently from a capsule, and differently again from a dried snack in a pouch.
| What is measured | Why it is on the list | Which SKUs it matters most for |
| Water activity and moisture | Sets whether the pack is shelf stable at ambient and whether it stays that way | Dried pieces, powders, tea, capsules |
| Microbiology: total plate count, yeast and mould, coliform | The safety and spoilage read that a listing submission asks for | Every ambient SKU, and the gating test for retort and hot-fill |
| Peroxide value and free fatty acid | Tracks oxidative rancidity in anything carrying fat or oil | Coconut products, nut-based spreads, granola, fried snacks |
| pH and titratable acidity | Confirms the acidification that the thermal process was designed around | Acidified jars, sauces, pickles, beverages |
| Colour, texture, and sensory panel | Catches the failure that is not unsafe but is unsellable | All retail SKUs, and the usual reason a date gets shortened |
| Pack integrity: seal strength and headspace | A shelf life is a property of the product inside the pack, not the product alone | Pouches, glass jars, capsule bottles |
Local capacity for this work exists. The Industrial Technology Institute lists shelf-life testing of final or packaged products by real-time and accelerated methods, sensory analysis through a trained taste panel maintained on site, and thermal process validation of canned and bottled foods. A Sri Lankan brand owner does not need to send samples offshore to build a defensible file.
How iteration counts and retort cycles interact on a formulation project is described in vegan cheese spread R&D in Sri Lanka: the retort iteration cycle.
How does the result gate a retail listing?
The stated shelf life is an input to everything downstream. SLSI clearance for a packaged food typically runs 4 to 8 weeks, so a project should hold a 6 to 10 week buffer between quality sign-off and the intended shelf date. Buyers at the major supermarket chains and the online marketplaces ask for a shelf life on the listing form, then set a minimum residual life at delivery that shortens the commercial window at the other end.
A pattern shows up often enough at the Matale facility to be worth naming. A founder arrives with a finished label design carrying a 24-month best-before, chosen because a competitor product on the same shelf shows 24 months. No study exists. Once the accelerated data comes back, the honest number is 9 or 12 months, and the choice is between reprinting a label order already paid for or reformulating and repacking to reach the printed claim. The reprint is the cheaper mistake, and it is entirely avoidable by settling the date before the artwork goes to press.
The reason to get this right is commercial as well as legal. Sri Lankan food and beverage exports rose 24.23 percent to USD 583.89 million in 2025, with processed food exports up 40.4 percent, according to the Sri Lanka Export Development Board. A local brand that later wants an export listing gets asked for the stability file, and the file built properly the first time is the one that travels.
The submission window itself, worked through on a real category, is in SLSI clearance for a private-label coconut jam.
Service snapshot: where the stability work sits at Silk Foods Ceylon
|
What does a stability brief need to state?
The R&D side logs roughly 40 first-brief inquiries a quarter, and close to 60 percent arrive without a stable formulation. A brief carrying the six lines below can be costed and scheduled in one pass, because each line removes a decision that would otherwise be taken twice.
Buyer's checklist: what a stability brief must state
|
None of those lines needs a laboratory to answer. All of them change the quote, and all of them are cheaper to settle in week one than in month nine.
What the surrounding iteration budget looks like for a first formulation project is set out in food product development in Sri Lanka: budgeting for R&D iterations.
Frequently asked questions
How long does shelf-life testing take for a Sri Lankan packaged food?
A full-term study takes as long as the claim, so 12 months of evidence supports a 12-month date. An accelerated run at 40 degrees Celsius and 75 percent relative humidity follows the ICH Q1A(R2) convention of 6 months with pulls at 0, 3, and 6, which gives an early read while the full-term study continues.
Can accelerated testing replace a full-term shelf-life study?
Not on its own. The Institute of Food Technologists noted in 2023 that adding 10 degrees and halving the time oversimplifies, because degradation mechanisms in one product respond differently to heat. Accelerated data reads oxidation and browning well, reads separation and moisture migration poorly, and supports a provisional date pending the full-term result.
Does Silk Foods Ceylon handle shelf-life studies within a contract manufacturing project?
Yes. The study is scoped inside the R&D and NPD stage at the Matale facility, which adds 4 to 6 weeks at the front of a project, with 2 to 4 sample iterations before a spec locks. Purchase order to dispatch then runs 2 to 3 weeks on an existing spec and 6 to 10 weeks where formulation comes first.
What does the 2026 labelling regulation require on the date of expiry?
The Food (Labelling and Advertising) Regulations 2026, in operation from 1 July 2026, require a numerical date of expiry preceded by Expiry, Use before, Use by or Best Before, printed at not less than 1.5 millimetres. Regulation 5(4) prohibits selling food after that date, and a month-and-year marking expires on the last day of that month.
How Silk Foods Ceylon can help
For founders and local brand owners taking a first SKU to a retail shelf, Silk Foods Ceylon runs the formulation, the stability scoping, and the commercial batch inside one project at the Matale facility, so the date on the label is settled before the artwork is printed. Formulation adds 4 to 6 weeks at the front, samples run 1 to 2 weeks on an existing recipe and 2 to 4 weeks per iteration on a new one, and purchase order to dispatch is 2 to 3 weeks on a locked spec. First-run MOQs are 1,500 jars on a 300 g glass spread, 1,250 bottles on a 200 ml beverage, and 180 bottles on a 60-count capsule SKU. The facility is BRCGS- and FSSC 22000 V6-audited, with SLSI submission support and Sri Lanka Food Act labelling compliance built into a standard engagement.
A brief that states the intended claim, the actual pack, and the distribution condition can be costed in a single pass.
To brief a project and request a feasibility call, email b2b@esilkroute.com.lk or call +94 76 441 0389 / +94 76 918 5744.
Sources
Food (Labelling and Advertising) Regulations 2026, Gazette Extraordinary No. 2494/47 of 26 June 2026, made under the Food Act, No. 26 of 1980, Ministry of Health. eohfs.health.gov.lk (retrieved 22 August 2026).
Food Act, No. 26 of 1980, section 32, Ministry of Health, Sri Lanka. eohfs.health.gov.lk (retrieved 22 August 2026).
Codex Alimentarius Commission, General Standard for the Labelling of Prepackaged Foods, CXS 1-1985, adopted 1985 and amended through 2010, on date of minimum durability and use-by definitions. fao.org (retrieved 22 August 2026).
International Council for Harmonisation, ICH Q1A(R2), Stability Testing of New Drug Substances and Products, 2003, on accelerated storage at 40 degrees Celsius and 75 percent relative humidity. database.ich.org (retrieved 22 August 2026).
World Health Organization, Annex 10, Stability testing of active pharmaceutical ingredients and finished pharmaceutical products, WHO Technical Report Series No. 1010, 2018, on climatic-zone storage conditions. database.ich.org (retrieved 22 August 2026).
Institute of Food Technologists, How to Fast Track Your Shelf Life Testing, Food Technology, September 2023, on the Q10 convention. ift.org (retrieved 22 August 2026).
Food and Drink Federation, Industry Guidance on Setting Product Shelf-Life, Version 1, November 2017. fdf.org.uk (retrieved 22 August 2026).
Industrial Technology Institute, Food Technology services, on shelf-life testing by real-time and accelerated methods. iti.lk (retrieved 22 August 2026).
Sri Lanka Export Development Board, Sri Lanka’s Export Performance Exceeded US$ 17.2 Billion in 2025, January 2026. srilankabusiness.com (retrieved 22 August 2026).
Written by the Silk Foods Ceylon Editorial Team. Silk Foods Ceylon (Pvt) Ltd. is a BRCGS- and FSSC 22000 V6-audited contract manufacturer in Matale, Sri Lanka, offering contract manufacturing, private labelling, co-packing, and in-house R&D for Sri Lankan brand owners, FMCG companies, hotel and restaurant groups, and distributors. To brief a project: b2b@esilkroute.com.lk, +94 76 441 0389, or +94 76 918 5744.