What it means
A buyer may approve a colour, material, flavour or component finish, but later production can drift, so both sides need to know what was agreed. Retain a sample with item code, supplier, lot, version, approval date, test status and storage conditions, and record exactly what the sample represents, because a reference for appearance may not prove performance.
Choose retention quantity and period by product risk, shelf life, contract and applicable regulation, remembering that some samples degrade in light, heat or humidity. Seal samples where necessary and record any opening or testing, and record the agreed comparison method alongside the retained item.
A digital photograph can support a physical sample but may not capture texture, exact colour or material properties reliably. Define ownership and access, since both buyer and supplier may hold matched references.
If one side changes a specification, retire the old version with a clear link rather than silently swapping the sample, so that when a dispute arises the current batch is compared with the right version under a repeatable method, not by memory. Keep samples separate from saleable stock and make disposal controlled, because a retained product with an expired date should not return to fulfilment.
Track location and retrieval time so a claim is not delayed while staff search across sites. For owners, retention is a low-cost form of evidence when goods vary across time, though it does not replace written specifications, incoming tests or contractual rights.
For a regulated product, the applicable agreement and rules may prescribe containers, storage and retention periods, so a period should not be copied from another industry. The owner should review whether the sample remains usable for its intended test at the end of storage.
When removing a sample for investigation, log who took it, when and how much was consumed, and keep enough evidence to reproduce a disputed result where practical. If storage conditions failed, flag the reference as compromised rather than quietly relying on its appearance.
Choose sample quantity based on the tests likely to be repeated, not simply the space available on a shelf, and if a test destroys the sample, plan how many units each party may need and which one is the arbitration reference. Label any difference between an approved prototype and routine production samples so the wrong standard is not used during any later quality dispute.
In practice
Real-world examples.
Example
A fabric buyer keeps a sealed swatch beside a signed colour specification for seasonal reorders. When a new roll arrives, staff compare it with the swatch under the same lighting. The swatch is returned to its sealed envelope and the check is logged.
Example
A food manufacturer retains a labelled unit from each defined batch under its product requirements. The unit is stored in the conditions the requirements specify. After the retention period ends, a controlled disposal record is completed.
Example
A supplier changes a material; both parties archive the old approved sample and label the new version clearly. The old sample is marked as superseded with the date of the change. Future disputes about older orders can still be checked against the right reference.
Formula
Calculation
Reference sample traceability rate = Required retained samples with complete identity and location records / Required samples selected for retention x 100
Worked example. An invented buyer requires 100 retained samples across current supplier batches. Ninety-four are correctly labelled and locatable, while six lack a batch ID or known location.
- Traceability rate = 94 / 100 x 100 = 94%.
- The six gaps limit the value of any later comparison.
Retention also has a running cost. If each retained sample costs $4 a year to store, label and log, then 100 samples cost 100 x $4 = $400 a year. Compared with a single disputed batch worth $12,000, the cost of reliable evidence is small, although the saving depends on whether the sample is actually usable when needed.
The required sample population depends on risk and agreements.Case study
Seen in the real world.
This illustrative and entirely fictional example follows Riverstone Textiles, an invented furniture maker. A later delivery of fabric looked darker than the approved prototype. Both sides claimed their version was correct, but the buyer's sample was unlabelled and had been left in sunlight. Riverstone and the supplier agreed on controlled reference swatches with version IDs, sealed storage and a colour-test method.
New deliveries are compared with the relevant reference, while older versions remain archived for order disputes. The next difference was resolved with a shared test rather than two faded memories. Riverstone also records when a swatch is removed for testing. A faded or damaged reference is not used to judge a supplier batch without qualification.
The team keeps the old swatch as evidence of what went wrong in storage, while a properly approved new reference governs future deliveries. Riverstone's purchasing manager now reviews the retention log each quarter, checking that every active fabric has a labelled reference, a storage location and a review date. The invented example shows a method for keeping evidence usable, not a claim about any saving.
Watch out
Common mistakes.
- Keeping a sample with no batch, approval version or storage record.
- Assuming a visual reference proves every performance requirement.
- Replacing an old reference silently after a specification change.
Questions
People also ask.
How long should a sample be kept?
Follow product risk, contract and applicable rules; set a documented period.
Is a photograph enough?
It can support some checks, but many physical properties need an actual controlled sample.
Can a retained sample be sold later?
Only if product condition and applicable rules permit; keep retained stock controlled until disposition.
From the founder's library
Take it further with the book.
Build your financial confidence beyond this definition. Shihan's full-length guide, Accounting Fundamentals, takes the same plain-English approach and turns it into a complete, practical playbook for non-finance managers, business owners and students - with chapter-end quiz answers and presentation slides included.
25% off with code MMHQ25, applied at checkout. Priced in USD - checkout may show the equivalent in your local currency.
View the book and save 25%