Bulk Olive Oil Sampling Protocol: ISO 5555 & IOC Method

A flawless CoA protects nobody if the sample behind it isn't the one actually loaded into the tank. Between the mill's storage tank and the container leaving Radès port, it's the sampling method — not the lab report — that separates admissible evidence from a piece of paper. Here's the protocol a bulk buyer should require: standardized vocabulary, a method matched to the container, sealed counter-samples, and a chain of custody for disputes.

A CoA is only as good as the sample behind it

A lab report describes a sample, not a lot. If that sample was taken badly — a single point near the surface of a poorly mixed tank, an unsealed bottle, an excessive delay reaching the lab — the CoA can be perfectly accurate and still say nothing about the oil actually delivered. The reference standard governing this step, too often treated as an afterthought, is ISO 5555: Animal and vegetable fats and oils — Sampling. First published in 2001 and reconfirmed in 2023, it sets out the vocabulary and reference methods that accredited labs and the International Olive Council build their own protocols on.

The standardized vocabulary: increment, bulk sample, laboratory sample

ISO 5555 draws three distinctions a buyer should be able to name precisely:

• Increment: the quantity of oil taken from one place, at one moment. A single increment, pulled from anywhere in a 200-tonne tank, carries no representative value on its own.
• Bulk sample: the combination of several increments, in proportion to the quantities they represent, meant to reflect the whole lot.
• Laboratory sample: the quantity obtained after homogenizing and reducing the bulk sample — the portion that actually goes for analysis.

The standard also addresses a detail buyers routinely overlook: for a fat that isn't perfectly homogeneous, increments should be drawn every 300 mm of depth, from the top of the tank down, plus a dedicated bottom sample. A 2014 amendment added an annex specifically for sampling flexitanks, acknowledging that this packaging format poses different access constraints than a fixed tank.

Sampling by container type

The method changes sharply depending on what the oil travels in. A single access point at the top of a flexitank doesn't allow the same protocol as a stainless tank fitted with sampling valves at three heights.

Container	Sampling method	What to watch for
Fixed stainless tank (mill)	Increments at several heights via dedicated valves, combined into a bulk sample	Mix or recirculate before sampling to break up any stratification
Flexitank	Drawn through the fill/drain valve, per the dedicated ISO 5555 annex	Single access point only: increase sampling frequency over time (start/mid/end of filling)
Isotank	Drawn through the top dome or bottom valve, depending on configuration	Check for residue from the previous cargo before the first increment
Drums, IBCs	One increment per container across a lot sample, probe for sealed drums	Hermetically sealed drums with sound internal lacquer are a precondition for a valid sample

The IOC's method for large tanks

For large tanks sampled through the top opening, the IOC's trade standard (COI/T.15/NC No 3) details a three-zone split: one part from the top 10% of the tank, three parts spread across the middle 30%, and one part from the bottom 10%. Everything is blended in equal proportions into one final sample. This weighting isn't arbitrary — it compensates for the fact that oil in a tank is never perfectly uniform in temperature, residual moisture, and sediment, even after mixing.

Sealed counter-samples and chain of custody

A sample without a seal is a statement of intent, nothing more. The standard we apply — and that any serious buyer should write into a contract — is this: every sample is drawn jointly by both parties, split into several identical bottles, sealed with a uniquely numbered tag recorded on a report signed by both sides. One bottle goes to the lab, a second stays sealed with the seller, and a third — the counter-sample — can be lodged with a neutral third party (freight forwarder, independent lab) to settle a disagreement without relying on either side's goodwill.

ISO 5555 adds a timing constraint that's easy to forget: the sample must reach the laboratory no later than the fifth working day after collection; past that point, it must be stored under conditions that prevent degradation before analysis. A sample sitting on a desk for three weeks before being shipped to the lab no longer describes the lot at the moment it was drawn — it describes its oxidation in the meantime.

Three sampling moments: loading, arrival, dispute

Three distinct moments call for separate samples, and conflating them is a classic source of commercial disagreement:

1. At loading, with both parties or their representatives present (forwarder, appointed inspector): this is the contractual reference sample.
2. On arrival, to verify that the delivered lot matches the reference sample; a significant discrepancy sends both sides straight back to the sealed bottles kept in storage.
3. In a dispute, the counter-sample held by the neutral third party is analyzed by an independent lab, whose report contractually prevails over the first two if the contract said so upfront.

A contract that never states which of the three governs in case of a discrepancy leaves the door open to a standoff with no referee. Better to fix it in writing before the first shipment.

A real-world case: resolving a gap between two samples

A buyer receives a container whose arrival CoA shows a peroxide value of 24 meq O₂/kg, against 14 on the loading bulletin. Without a sealed counter-sample, the file ends there — a good-faith argument with no admissible proof of what actually happened in transit. With a sealed bottle lodged at a neutral third party at loading, the process changes shape: that bottle goes to an independent lab, and its result settles the question. If it confirms a value close to 14, the problem lies in transport or storage at destination — heat, delay, a break in the handling conditions bulk edible oil requires. If it already confirms a value near 24, the lot was already degrading at loading, and liability sits with the seller. Either way, it's the sampling protocol — not the commercial back-and-forth — that produced the answer.

Mistakes that invalidate a sample

• A "representative" sample pulled from a showroom stock, unconnected to the lot actually loaded: it carries no contractual value, however good the accompanying bulletin looks.
• An unsealed bottle or an untagged seal: nothing then rules out a substitution, intentional or not, between sampling and analysis.
• An unmixed tank before sampling: oil stratifies slightly over time; a surface-only increment systematically overstates the freshness of the lot.
• An uncontrolled transit delay: past the fifth working day without proper storage conditions, the sample has moved on, and the bulletin it produces describes a different oil than the one drawn on day one.
• No signed report: without a dated, jointly signed document, a sealed sample loses much of its evidentiary weight in a dispute.

What Virginia applies to every lot

At Virginia, a jointly sealed sample isn't a commercial add-on — it's the baseline condition of every shipment: systematic tasting, a per-lot CoA covering acidity, peroxide, K232/K270 and polyphenols on request, traceability back to the partner mill, and independent SGS counter-analysis accepted at loading. Our lots come from a network of partner mills in Tunisia, held in stainless tanks under nitrogen at those same partner sites before shipment — nothing that can't be checked against documentation. Request sealed samples of our oils to test the protocol yourself on a real lot, or request a quote specifying your own sampling requirements — qualified response within 24 business hours. For more on reading the results, see our guide to reading a CoA, the ten points for choosing a bulk supplier, and our flexitank vs. isotank comparison to match the protocol to your mode of transport.