How to Take a Representative Sample from a Bulk Tea Lot

A representative bulk tea sample is built from controlled increments taken across one clearly defined lot, then combined, reduced, divided, sealed and recorded without favouring the easiest bags or the best-looking leaf. A grab sample from an open bag may be useful for a quick look, but it cannot support a shipment-wide acceptance decision unless the contract expressly makes that limited sample controlling.

Define the lot before touching the tea

The first sampling decision is not how much tea to scoop. It is what population the result will represent. Write the supplier lot or production code, tea type and grade, pack type, number of packages, net mass, warehouse location and current status. Freeze that population for the sampling event: packages added, removed, reworked or relabelled afterwards break the link between sample and shipment.

Do not treat different production batches, grades, pack sizes or visibly different conditions as one homogeneous lot merely because they will travel in the same container. Codex's revised 2023 General Guidelines on Sampling, CXG 50-2004, warns that applying ordinary plans to inhomogeneous lots can increase both producer and consumer risks. Where practical, separate the inspection lot into defensible production sublots before selecting samples.

Choose the controlling sampling reference

The ISO catalogue checked on 5 August 2026 lists ISO 1839:1980, Tea - Sampling as current after confirmation in 2024. Its scope applies to containers of all sizes and covers apparatus, sampling from containers, sample packing, labelling, dispatch and the sampling report. Obtain the standard and state its edition when the parties want it to control; a catalogue abstract is not a substitute for the procedure.

Keep five requirement types separate

• Trade reference: a grade code, approved sample or customary practice helps describe the tea but does not create a complete sampling rule.
• Legal requirement: a mandatory destination rule may prescribe a method for a specific control. For example, Commission Directive 2002/63/EC establishes official-control sampling methods for pesticide residues in relevant foods; a commercial pre-shipment sample is not automatically an official-control sample.
• Voluntary standard: ISO 1839 and Codex guidance are technical references unless law or another binding instrument incorporates them.
• Buyer specification: the buyer states the lot, purpose, selection plan, sample quantities, handling and acceptance rule.
• Contract requirement: both parties agree which document and edition governs, who may sample, and what happens after a disputed result.

Use this seven-step field method

1. State the purpose. Identify whether the sample is for sensory comparison, moisture and composition, pesticide residues, microbiology, foreign matter or another test. Ask the laboratory for its required sample mass, container, preservation and transport conditions before collection. One handling method may not suit every analysis.
2. Lock the lot list. Record every eligible package or pallet and mark damaged, wet, opened or mismatched units. Investigate exceptions separately; do not quietly exclude them to improve the composite.
3. Select across the whole lot. Use documented random selection, or systematic selection with a random starting point, across pallet positions, rows and accessible levels. Codex says each discrete item should have an equal chance under random sampling and accepts systematic sampling in many cases. Its guidance for layered lots selects through pallet, outer carton, inner carton and item levels instead of sampling only the front row.
4. Collect increments consistently. Use clean, dry equipment suitable for tea and the intended analysis. Follow the agreed standard for access point, penetration and increment size. Prevent contamination, moisture gain, leaf breakage and loss of fines; tea can segregate by particle size during handling, so a convenient surface scoop can bias appearance and composition.
5. Build the aggregate sample. Combine the planned increments only when the decision concerns the lot average and the method permits compositing. Mix and reduce the aggregate using the agreed procedure. Do not remove attractive leaf for the display sample or discard dust before division.
6. Create controlled splits. Divide the reduced material into the laboratory sample and any buyer, supplier or referee samples required by contract. Use suitable clean containers, close them promptly and apply unique tamper-evident seals. Equivalent splits are useful only when they come from the same correctly mixed aggregate.
7. Complete the record. Record sampler, witnesses, date and time, location, lot and package list, selection method, packages opened, increment count and approximate mass, equipment, composite and reduction method, split IDs, seal numbers, storage, dispatch, deviations, photos and signatures.

Copy this sampling instruction into the purchase file

There is no defensible universal rule such as "sample five bags" for every tea shipment. The required scale depends on lot size and structure, the characteristic being controlled, destination law, method, acceptable risks and contract. If those inputs are not fixed, a larger handful does not repair the plan.

Know when a composite sample hides the problem

A composite can be efficient for estimating an average characteristic, but mixing dilutes local variation. Keep increments or package samples separate when the decision concerns package-to-package consistency, damage, odour, contamination hot spots, label differences or the location of a failure. The revised Codex guideline also distinguishes sampling uncertainty - sources before the laboratory sample arrives, including the procedure, lot variation, sampler and subsampling - from analytical measurement uncertainty. A precise laboratory result cannot correct a biased field sample.

For pesticide-residue work, first identify the destination's applicable legal sampling rule and laboratory scope. The EU's SAMANCTA representative-sampling guidance explains the core logic: increments should come from different places throughout the consignment and the final sample is created from their aggregate. Use our EU tea MRL pre-sampling checklist to define the analytes, legal snapshot and decision rule before collection.

Common buyer mistakes

• Sampling before defining and freezing the lot.
• Allowing the warehouse to present only clean, reachable or already opened bags.
• Using one surface grab to release a full lot.
• Combining different production batches without a documented reason.
• Discarding fines, stalk or broken leaf while making the composite look better.
• Sending an unsealed sample with only a product name and no lot ID.
• Testing a display or pre-contract sample as if it represented later production.
• Keeping no referee sample or contract procedure for a disputed result.
• Confusing a laboratory's analytical competence with proof that the field sample was representative.

Practical conclusion

Release evidence should form one uninterrupted chain: defined lot - documented selection - controlled increments - justified composite or separate samples - unbiased reduction - sealed splits - chain of custody - agreed test and decision rule. If any link is missing, classify the result as "clarify" rather than extending it to the shipment by assumption.

Use the site's bulk green tea purchase specification template to make the sampling plan contractual, compare it with Yunjing Tea's sample-to-shipment quality controls, and send the destination, tea type, lot structure and proposed test panel before sampling.

Sources checked 5 August 2026: ISO 1839:1980, Tea - Sampling (confirmed 2024); Codex CXG 50-2004, General Guidelines on Sampling (revised 2023); Commission Directive 2002/63/EC; and European Commission SAMANCTA representative-sampling guidance.