How to Verify Whether a Fine Robusta Supplier Actually Reduces Defect Risk

Updated: 4 September 2026

Primary keyword: Fine Robusta supplier defect risk verification

A supplier can claim better sorting, cleaner lots, lower defects and stronger consistency.

Those claims should be measured.

This page does not present an unverifiable case study about a specific UK roastery reducing defects by 42%. Instead, it gives buyers a repeatable method for determining whether a Fine Robusta supplier actually reduces operational risk.

Start with a baseline

Before changing suppliers, record the current condition.

A useful baseline can include:

• green defects observed per sample;
• moisture;
• water activity where available;
• foreign matter;
• odor issues;
• size distribution;
• roast loss;
• hand-sorting time;
• rejected batches;
• cupping defects;
• espresso dial-in waste;
• complaint rate;
• delivery delays caused by QC.

Without a baseline, improvement claims are difficult to prove.

The baseline should be tied to a defined period and sample method.

Define the sample method

Do not compare one carefully selected sample from Supplier A with a random production sample from Supplier B.

Use comparable sampling.

Record:

• sample weight;
• lot identity;
• where the sample was taken;
• whether it is offer, pre-shipment or arrival sample;
• date;
• packaging condition;
• evaluator.

Consistency of sampling is essential.

Measure physical defects with a named method

A supplier may describe coffee as “clean” or “premium.”

Translate that into observable evidence.

Depending on the contract, record:

• primary defects;
• secondary defects;
• insect damage;
• broken beans;
• foreign material;
• mold or odor concerns;
• black or sour beans where relevant;
• screen distribution.

If a historical protocol is used, identify it.

Do not claim that one defect threshold universally defines all current Fine Robusta transactions.

The buyer’s contract should state the method and tolerance.

Track moisture and storage stability

Moisture can affect storage, roast behavior and defect risk.

Water activity can add useful information where available.

Track:

• moisture at offer;
• moisture before shipment;
• moisture at arrival;
• warehouse condition;
• packaging;
• storage duration.

The important question is not only whether the number falls inside a generic internet range.

The important question is whether the coffee remains stable and fit for the intended use.

Compare offer, pre-shipment and arrival samples

This is one of the strongest controls.

For each stage, keep:

• physical results;
• sensory notes;
• sample identity;
• date;
• roast preparation;
• decision.

Then compare.

If the offer sample is clean but the arrival sample shows more defects or a different sensory profile, the evidence supports a supplier discussion.

Without this chain, teams often rely on memory.

Measure sorting loss

If the buyer performs additional sorting, record the removed mass.

For example:

• inbound sample weight;
• mass removed during hand sorting;
• reason for removal;
• final usable mass.

The percentage can then be calculated from actual data.

Do not use a generic industry statement such as “better suppliers reduce sorting by 60%.”

Your operation should generate its own number.

Measure roast loss separately

Roast loss and green sorting loss are different.

Record:

• green input mass;
• roasted output mass;
• roast profile;
• batch size;
• rejected roasted coffee.

A supplier may reduce green defects without changing normal roast mass loss.

Keep the metrics separate so improvement is not overstated.

Track sensory consistency

Supplier performance is not only physical.

Use repeat sensory evaluations.

Record:

• descriptive profile;
• sweetness;
• bitterness quality;
• acidity;
• body;
• aftertaste;
• defects;
• cup-to-cup variation;
• lot-to-lot variation.

The current SCA Coffee Value Assessment framework is useful conceptually because it separates descriptive, affective, physical and extrinsic information.

That helps a buyer distinguish “different” from “worse.”

Test the intended beverage

A lot can look physically strong and still underperform in the product.

For espresso, record:

• dose;
• yield;
• time;
• workable grind range;
• sweetness;
• bitterness;
• body;
• milk performance;
• dial-in waste.

For filter, record:

• clarity;
• sweetness;
• bitterness;
• cooling behavior;
• repeatability.

For cold brew or another format, use the real production method.

Supplier quality should be judged in the application that creates revenue.

Track rejection and corrective-action rates

A supplier-quality system needs decision outcomes.

For each shipment or lot, classify:

• accepted;
• accepted with deviation;
• rejected;
• reworked;
• replaced.

Then record:

• reason;
• corrective action;
• supplier response time;
• replacement time;
• commercial resolution.

Over time, these records create a supplier-performance history.

Measure delivery reliability

Quality can fail even when the cup is good.

Track:

• promised ship date;
• actual ship date;
• promised arrival;
• actual arrival;
• missing documents;
• packaging damage;
• customs or logistics issues attributable to documentation.

A supplier that consistently delivers the right coffee on time may create more value than a supplier with a more exciting sample and weak execution.

Separate certification from current quality

A historical certificate can support trust.

It does not replace current QC.

Coffee Quality Institute states that it stopped accepting coffee samples for grading on 30 September 2025. SCA launched the updated Q Grader program under SCA management on 1 October 2025.

Historical Q Robusta records should therefore be labeled as historical evidence.

The current shipment still needs current evidence.

Do not use screen size as a universal quality ranking

Screen size can matter operationally.

It can influence roast uniformity and contract specifications.

But larger beans are not automatically higher quality.

A supplier should not be judged only by screen number.

Use screen distribution where it matters, but keep it separate from sensory quality.

Build a supplier scorecard

A practical scorecard can include:

Identity

• lot clarity;
• harvest clarity;
• producer clarity.

Physical

• defects;
• moisture;
• water activity;
• packaging.

Sensory

• cleanliness;
• descriptive consistency;
• defects;
• application fit.

Commercial

• price;
• quantity;
• terms;
• lead time.

Reliability

• sample continuity;
• delivery;
• documentation;
• corrective action.

The weighting should match the buyer’s business.

A simple before-and-after design

To determine whether a new supplier improves performance:

1. define baseline period;
2. define comparable sample method;
3. collect at least several lots or shipments where possible;
4. measure the same variables;
5. calculate changes;
6. note confounding factors;
7. avoid claiming causation from one shipment.

For example, a lower defect rate may reflect better sorting.

It may also reflect a different harvest, different sample method or a smaller lot.

The buyer should document those differences.

Why one case study is not enough

A single success story can be useful.

But it should include verifiable source data.

Questions include:

• Is the roastery named?
• Is the supplier named?
• Are sample sizes available?
• Were defect methods consistent?
• Was the before period comparable?
• Was the same origin used?
• Was the same harvest used?
• Were logistics conditions similar?

If those details are missing, the case study should not be presented as measured fact.

That is why OCC uses a measurement framework here instead of an invented percentage.

What this means for Cambodia Fine Robusta

Cambodian suppliers can build trust by making verification easy.

Useful evidence includes:

• lot identity;
• producer identity;
• harvest;
• process;
• current physical data;
• current sensory data;
• sample continuity;
• storage;
• delivery documentation.

For broad supplier evaluation in Cambodia, use Evaluating Cambodian Coffee Suppliers.

Frequently asked questions

What is the best defect target?

There is no universal target for every contract. Use a named method and buyer-defined tolerance.

Does a certificate guarantee low defects?

No.

Does larger screen size guarantee quality?

No.

How many shipments are needed before calling a supplier consistent?

More than one. The more variable the product and season, the more repeated evidence is useful.

Should sensory data be included in supplier performance?

Yes.

Can a lower-priced supplier still be better?

Yes, if total delivered value, consistency and risk are better.

Bottom line

A supplier reduces defect risk only when the data show it.

Measure:

baseline → comparable samples → physical defects → sensory consistency → application performance → delivery → corrective action.

Do not replace measurement with an unsupported case-study percentage.

Sources

• Specialty Coffee Association, Coffee Value Assessment: https://sca.coffee/value-assessment
• Specialty Coffee Association, Coffee Standards: https://sca.coffee/research/coffee-standards
• Coffee Quality Institute, public coffee database: https://database.coffeeinstitute.org/coffees

Add one more control: verify the corrective-action process

Defect risk is not only about preventing every problem. It is also about how quickly a supplier can identify and correct a problem when it appears.

A buyer should record:

• who receives the complaint;
• how the supplier verifies the issue;
• whether retained samples are available;
• how quickly the supplier responds;
• whether the cause is identified;
• whether replacement, credit or another remedy is documented;
• whether the same problem appears in later shipments.

A supplier with transparent corrective action can be safer than one that claims perfect quality but cannot investigate deviations.

Corrective-action evidence also helps distinguish a one-time agricultural variation from a recurring process-control problem.

Keep the denominator visible

Whenever a defect percentage is reported, preserve the denominator.

“Defects fell by 40%” is incomplete unless the reader knows:

• the starting rate;
• the ending rate;
• sample size;
• sample method;
• number of lots;
• time period.

For example, moving from 5 defects to 3 defects in one small sample is mathematically a 40% reduction, but it does not automatically prove a stable supplier improvement.

Good supplier analysis reports both the percentage and the underlying counts.

That is the standard OCC should use when discussing procurement performance.