Sensory & Quality · Calibration

Use calibration to understand variation, not hide it.

Calibration is more than reaching the same score. It is a structured process for understanding agreement, repeatability, reference use and the differences that matter for a professional decision.

Compare
Independent observations
Diagnose
Where variation comes from
Repeat
Measure whether alignment improves

Calibration framework

A repeatable calibration cycle.

Useful calibration identifies the source of variation before deciding what kind of training, reference work or procedural change is appropriate.

01
Protocol

Standardize what can reasonably be standardized.

Sample preparation, presentation, timing and evaluation procedures should reduce avoidable variation before evaluator differences are interpreted.

02
References

Use physical references when language is not enough.

Reference materials can support recognition and intensity alignment when they are prepared consistently and matched to a defined calibration objective.

03
Repeatability

Ask whether evaluators can reproduce their own judgement.

Repeated presentations help separate within-evaluator inconsistency from meaningful differences between evaluators.

04
Decision

Calibrate for the decision the panel needs to make.

The required degree of alignment depends on whether the task concerns training, purchasing, production QC, descriptive work or another professional decision.

Current CVA assessment logic

Separate descriptive intensity from affective quality impression.

Descriptive Assessment records seven sensory-section intensities on a 0–15 scale without producing a quality total. Affective Assessment separately records impression of quality across eight sections on a 1–9 scale and converts those ratings to the SCA 104 cupping score.

Method basis: SCA Standard 103-2024 Coffee Value Assessment: Descriptive Assessment and SCA Standard 104-2024 Coffee Value Assessment: Affective Assessment.

Coffee Value Assessment

Two assessment purposes. Two different outputs.

Descriptive assessment records what is perceived and at what intensity. Affective assessment records a taster's impression of quality. The two measurements are intentionally kept separate here.

SCA Standard 103-2024

Descriptive Assessment

No quality total

Rate the perceived intensity of each sensory section on the 0–15 descriptive scale. Intensity is not a judgement of desirability or quality.

7.5
8.0
8.5
7.0
6.5
5.5
7.0

This web demonstrator uses 0.1 increments for interaction. The governing method describes a continuous 0–15 intensity line scale. Descriptor/CATA recording belongs with the full descriptive assessment and is not reproduced as an official form here.

SCA Standard 104-2024

Affective Assessment

Cupping score 79.00

Rate impression of quality from 1 (extremely low) to 9 (extremely high). A rating of 5 is neutral. The eight ratings are converted to the affective cupping score before non-uniform and defective-cup deductions.

5
5
5
5
5
5
5
5
Uniformity and defect deductions

−2 points per non-uniform cup.

−4 points per defective cup.

Transparent score calculation

S = 0.65625 × Σhᵢ + 52.75− 2u − 4d

Section-rating sum:40· Rounded to the nearest 0.25 point.

Method basis: SCA Standard 103-2024 — Descriptive Assessment and SCA Standard 104-2024 — Affective Assessment.

This is an educational interaction model, not an official SCA form and not a recorded result from a real coffee sample.

Continue the reasoning

Agreement and consistency are different questions.

A single session can reveal differences, but reliable sensory practice depends on repeated evaluation and a clear distinction between within-evaluator consistency and panel alignment.

Read What Does Sensory Calibration Actually Mean in Coffee? and Repeatability vs Reproducibility in Coffee Sensory Evaluation.

Calibration in practice

A calibrated panel is useful when its observations support better decisions.

SZ Coffee can connect calibration with sensory education, evaluation systems and operational quality control.

Discuss calibration