A desiccated coconut COA has exactly six numbers that carry international standing. Everything else on the page is either a buyer specification or a courtesy. The six come from Codex Alimentarius CXS 177-1991, the Standard for Desiccated Coconut: acidity of the extracted oil, moisture, oil content, ash, extraneous vegetable material, and foreign matter.
A certificate of analysis (COA) is the lab report a supplier issues against a shipped lot. It is what your QA team accepts or rejects the container on, and the first document an auditor pulls.
Here is the part that costs money. The two parameters that actually get containers detained at destination, sulphur dioxide and Salmonella, appear nowhere in those six. They sit in destination-market law, and that law differs sharply between the EU, the US and Australia. A COA that clears Codex can still fail at the port.
What does Codex require on a desiccated coconut COA?
Six values. The limits run tighter than most buyers assume.
| Parameter | Codex CXS 177-1991 limit |
|---|---|
| Acidity of extracted oil (as lauric acid) | max 0.3% m/m |
| Moisture | max 4% m/m |
| Oil content, full-fat | min 60% m/m |
| Oil content, partially defatted | 35% to under 60% m/m |
| Ash | max 2.5% m/m |
| Extraneous vegetable material | max 15 fragments per 100 g |
| Foreign matter | absent in 100 g |
The acidity line is the one most often misread. Codex measures it on the oil extracted from the flake, expressed as lauric acid, not on the coconut as received. A COA reporting “FFA 0.15%” without stating the basis is ambiguous. Converted to an acid value, 0.3% as lauric acid works out near 0.84 mg KOH/g. Codex allows virgin coconut oil up to 4.0 mg KOH/g under CXS 210-1999. The desiccated coconut standard polices the oil inside the flake roughly five times harder than the oils standard polices bottled virgin oil.
Ask for the basis in writing. Add it to the spec sheet, not just the purchase order.
Which numbers on your COA are probably wrong?
Three figures circulate widely and none of them are in the Codex text.
- “Moisture max 3% per Codex.” Codex says 4%. The 3% figure is a commercial market expectation, and CBI lists it as a European buyer requirement. It is a real bar, but it is a buyer bar, not a legal one.
- “Fat min 55% per Codex.” Codex says 60% minimum for full-fat. A lot arriving at 57% fat is off-spec against the standard, whatever the supplier’s template says.
- “Acid-insoluble ash max 1.3%.” No such parameter exists in CXS 177. Codex sets total ash at 2.5% and stops.
The practical effect is a paperwork mismatch. Your specification cites Codex, the supplier’s COA cites a figure that is not, and neither side notices until a lot sits between the two. The desk writes both into the contract: the Codex limit as the legal floor, the commercial spec as the accepted one. Our own desiccated coconut programme runs moisture at max 3% and fat at min 65%, tighter than Codex on both.
Why does moisture matter beyond the number?
Moisture is a shelf-life parameter, not a safety parameter, and buyers routinely treat it as both.
Salmonella needs a minimum water activity of 0.93 to grow, per the FSANZ imported food risk statement on dried coconut. Desiccated coconut sits far below that, so the pathogen cannot multiply in the bag. It also does not die. The same document records Salmonella surviving months to years in low-moisture foods, and notes that high-fat, low-moisture matrices protect the organism against heat.
That combination is why a clean moisture figure tells you nothing about microbiological safety. A 2.8% moisture lot is a stable lot. It is not a validated-kill-step lot.
The validated control is upstream. FSANZ cites water-bath pasteurisation of raw coconut meat at 80 °C for 8 to 10 minutes as effective against Salmonella. Ask for the kill-step record, not the moisture number.
Where do SO2 and Salmonella actually sit?
Outside Codex, in three different regulatory shapes.
| Market | Parameter | Requirement | Nature of the rule |
|---|---|---|---|
| EU | Sulphites (as SO2) | below 50 mg/kg | Compositional cap on the product |
| US | Sulphiting agents | 10 ppm triggers label declaration | Labelling threshold, not a safety cap |
| Australia / NZ | Salmonella | n=10, c=0, not detected in 25 g | Binding sampling plan |
| EU | Salmonella | no coconut-specific criterion | Buyer specification governs |
CXS 177 names exactly one permitted additive, citric acid (INS 330), at GMP, and defers everything else to the General Standard for Food Additives. It sets no SO2 limit of its own.
The EU cap is the operative one for most programmes. CBI records Indonesian desiccated coconut pulled from the Polish market in January 2020 at roughly double the sulphite limit. The US number looks stricter at 10 ppm, but it is a different instrument: 10 ppm or more under 21 CFR 101.100(a)(4) means you declare sulphites on the label. It does not make the lot illegal. A buyer selling clean-label product into a US retailer treats 10 ppm as a hard reject anyway.
Salmonella is the asymmetry worth internalising. Regulation (EC) 2073/2005 Annex I Chapter 1 contains no row for desiccated coconut, dried fruit, or low-moisture ready-to-eat foods. There is no EU-wide statutory sampling plan for this commodity. Australia does have one, and it is severe: Schedule 27 of the ANZ Food Standards Code, in force 18 June 2025, sets n=10 for dried, chipped and desiccated coconut, against n=5 for pepper, paprika, cinnamon and cocoa powder. Regulators treat coconut as double the sampling risk of spices.
For scale, the Australian Imported Food Inspection Scheme ran 870 Salmonella tests on dried coconut from January 2007 to May 2016: 11 failures, a 1.3% rate. The US maintains Import Alert 23-12, detention without physical examination of coconut for microbiological contamination.
Does the COA state its test method?
If it does not, the number is not comparable to anything.
Codex assigns Type I status to four of the six methods, meaning the limit is defined by that method. Change the method, change the result.
| Parameter | Codex-referenced method |
|---|---|
| Moisture | AOAC 925.40 |
| Oil content | AOAC 948.22 |
| Ash | AOAC 950.49 |
| Acidity of extracted oil | ISO 660:2009 or AOCS Cd 3d-63 |
| Extraneous vegetable material | Visual count, 100 g thin layer on white background |
| Salmonella (where criteria apply) | EN ISO 6579-1 |
A COA listing “Moisture 3.1%” with no method is a claim. A COA listing “Moisture 3.1%, AOAC 925.40” is a result. When lots from two origins are being compared on the same spec, that distinction decides whether the comparison holds.
Peroxide value sits in a similar grey zone. Codex sets none for desiccated coconut. The 10 meq/kg and 15 meq/kg figures in CXS 210 apply to coconut oil, and that standard calls them supplementary factors, not conditions of conformity. Specify PV as a buyer requirement or not at all.
What this means for buyers
- Procurement (CPG): write both numbers into the spec, the Codex limit and your commercial limit, with the acidity basis stated as “extracted oil, as lauric acid”. Reject COAs without method references.
- Procurement (CPG): moisture is not a safety parameter. Request the kill-step validation record separately, and reference the Codex hygiene codes CAC/RCP 4-1971 and CAC/RCP 75-2015 in the supplier agreement.
- Importers and distributors: SO2 is destination-specific. A lot cleared for the US at 8 ppm still needs to sit below 50 mg/kg for the EU, and a clean-label programme needs neither.
- Brand owners: if any volume is destined for Australia or New Zealand, budget for n=10 sampling and confirm the supplier’s lab can support it before contracting.
- Traders: the Codex-versus-commercial gap on moisture and fat is a routine source of claims.
FAQ
What is a COA in coconut trading? A certificate of analysis is the lab report issued against a specific shipped lot, covering the chemical, physical and microbiological parameters agreed in the contract. It is lot-specific, unlike a technical data sheet, which describes the product in general.
Does Codex set a Salmonella limit for desiccated coconut? No. CXS 177-1991 sets no microbiological criteria. Salmonella requirements come from destination-market law or the buyer’s own specification. Australia and New Zealand apply n=10, c=0, absence in 25 g; the EU has no coconut-specific criterion.
Is desiccated coconut allowed to contain sulphur dioxide? Codex permits only citric acid by name and defers additives to the General Standard for Food Additives. For the EU market, sulphites must stay below 50 mg/kg. In the US, 10 ppm or more triggers a label declaration.
What FFA level should I accept? Codex caps acidity of the extracted oil at 0.3% m/m as lauric acid. Confirm the COA states that basis. A figure quoted on the whole product rather than the extracted oil is not comparable to the Codex limit.
Why do supplier COAs quote moisture at 3%? Because 3% is the European commercial expectation, not the Codex limit of 4%. Both are legitimate. Problems start when the contract cites one and the COA reports the other.
Locking one COA format across four origins is the harder half of the job. Send the desk your spec and we will tell you which origins can hold it.