03 · Basic · Intermediate
Sensory — Turning Perception into Language and Score
Cupping, brewed and espresso evaluation protocols and their scoresheets
Written by Chanho Hong
What this chapter covers
The SCA cupping protocol and how to read a cupping form; how expectations differ by processing method; the WBrC brewed tasting and WBC espresso tasting protocols; and what each scoresheet line is really asking of the taster.
Sensory evaluation is training that converts perception into reproducible data. Quality can only be discussed once several people are able to evaluate the same coffee in a similar way (reliability). This chapter covers the protocols and scoring systems of three standard evaluations — cupping (green coffee and roast evaluation), brewing tasting (WBrC) and espresso tasting (WBC).
Cupping — Comparing Coffees under Standard Conditions
Cupping is a method of evaluating and comparing several coffees side by side under standardised extraction conditions. Fixing the variables (same ratio, water, temperature and time) exposes only the differences in the coffee itself. It is the common language of green coffee buying, quality control and roast evaluation.
Protocol — The Cupping Protocol
| Item | Standard value | Rationale |
|---|---|---|
| Roast freshness | 8–24 hours after roasting | Minimises degassing variation; unifies conditions |
| Roast colour | Agtron #58 (whole bean) / #63 (ground) | Removes roast bias (light to medium) |
| Ratio | 8.25 g : 150 mL (≈1:18.18) | Standard strength |
| Water | 125–175 ppm, 93℃ | Appropriate minerals and temperature |
| Break | Stir the surface at 4 minutes to release the aroma | Unifies the timing of Aroma assessment |
Cupping Form — What Is Scored
The traditional SCAA form scored ten attributes. The current SCA CVA (Coffee Value Assessment) restructures these into Descriptive (intensity and descriptors) + Affective (seven quality attributes).
| SCAA ten attributes (reference) | CVA seven attributes (current canonical) |
|---|---|
| Fragrance/Aroma | Aroma (Fragrance is recorded separately as a descriptor) |
| Flavor | Flavor |
| Aftertaste | Aftertaste |
| Acidity | Acidity |
| Body | Mouthfeel (broken down into thickness + texture) |
| Sweetness | Sweetness |
| Balance / Clean Cup / Uniformity | Consolidated into Overall (Uniformity and defects are recorded separately) |
| Overall | Overall |
0 = cannot be assessed, 5 = neither high nor low, 9 = extremely high. Half points are not permitted, and 1–3 are not recommended in competition. This scale records intensity; the quality impression is assessed on a separate axis. The specialty baseline is 80 points or above (converted to the Legacy 100-point scale).
Evaluation by Processing — Washed · Natural · Others
Cleanliness, transparency and the structure of the acidity are central. What is examined is how clearly and how faultlessly the origin and varietal character come through. Fermented notes and coarse off-flavours are grounds for deduction.
Fruit, sweetness and body are emphasised, so the key question is whether that intensity maintains cleanliness and balance. Fermented nuance may be character, but it must be distinguished from defects such as over-ripeness, vinegar and mould.
Assessing Others (experimental processing) — with anaerobic fermentation, carbonic maceration and the like, the processing leads the flavour. The boundary between "defect" and "intended character" is ambiguous, so record the intensity (Descriptive) exactly as it is and judge the quality (Affective) by balance, cleanliness and whether the aftertaste is pleasant or unpleasant.
① Deciding on a green coffee purchase. Cup three lots side by side in the same session. If "Natural A is showy in fruit and sweetness, but the fermented note runs unpleasantly long in the aftertaste (Overall deduction)" and "Washed B is moderate in intensity but stable in cleanliness and balance", the choice diverges according to the intended use (blend vs single origin). The decision is made on the attribute-by-attribute profile, not the total score.
② Roast batch QC. Cup yesterday's and today's batches of the same roasted coffee side by side and look at Uniformity. If yesterday was "sweetness 6" and today is "sweetness 4, grassy", that is a signal of batch variation, so cross-check the colour and DTR records and correct the roast. Cupping becomes the measuring instrument of batch-to-batch reproducibility.
③ Calibration (aligning assessors). Several people score the same coffee and then compare scores and descriptors. If A gives "acidity 8" and B gives "acidity 5", they check with one another whether intensity has been confused with quality (intensity ≠ quality) and align the evaluation language. This is the precondition for reliable QC.
Brewing Tasting — The WBrC Protocol
Where cupping is "fix the conditions, compare the coffees", brewing tasting evaluates a single cup that the barista has optimised. The evaluation system of the World Brewers Cup (WBrC) is the standard for this.
WBrC Cup Score (seven attributes) evaluates brewed coffee on seven axes — Aroma · Flavor · Aftertaste · Acidity · Sweetness · Mouthfeel · Overall — each on a 0–9 scale. It uses the same vocabulary and scale as the CVA.
| Attribute | Definition |
|---|---|
| Aroma | The aroma of the brew. Assessed in the order intensity → complexity and clarity → descriptors |
| Flavor | The basic tastes (sweet, sour, salty, bitter, umami) combined with retronasal aroma. The "mid-range" that stays in the mouth |
| Aftertaste | The taste and aroma remaining after swallowing or spitting. Positive length earns points; negative persistence (lingering bitterness) loses them |
| Acidity | The perception of acid. Pleasant, it is brightness; unpleasant, it is sour. Intensity has no bearing on the score |
| Sweetness | The impression of sweetness. Intensity and quality are separate |
| Mouthfeel | The tactile sense of thickness plus texture. Not Flavor. Independent of intensity |
| Overall | Integrated assessment + Balance + Stability (holding character as the cup cools) |
In the Open Service, Barista Evaluation is added on top of this: descriptor accuracy (Accuracy, ×2), Attention to Details, coffee knowledge and equipment use, and presentation. The head judge separately assesses the Technical Uniformity (consistency of ratio, grind size, time, temperature and turbulence) across the three cups.
Scoresheet — Score Structure and Scales
| Scale | Range | Half points | Use |
|---|---|---|---|
| Coffee Evaluation (CVA) | 0–9 | Not permitted | The seven cup attributes |
| Accuracy | 0–3 | Not permitted | Descriptor accuracy |
| Impression | 0–3 | Not permitted | Visual and tidiness impression |
| Experience | 0–6 | Permitted (1–6) | Quality of the sensory experience |
① Purpose: cupping compares several coffees (conditions fixed); brewing evaluates the result of optimisation in a single cup. ② Variables: cupping fixes the extraction variables, whereas in brewing the barista freely designs ratio, grind size, water, temperature and technique. ③ Additional axes: brewing also scores the barista's communication, accuracy and consistency. ④ Even so, the vocabulary and scale of the cup score (CVA seven attributes, 0–9) are identical.
① Self-diagnosis of a home brew. If a cup is recorded on the CVA seven attributes as "Acidity 7 · Sweetness 5 · Mouthfeel 4 (thin) · Aftertaste 4 (short) · Overall 5", the profile has intensity but is weak in sweetness, body and finish. The diagnosis points towards under-extraction, and with the Chapter 02 mapping you adjust to a finer grind and a shorter ratio. The evaluation becomes the starting point of the next recipe.
② Tuning a competition routine. When preparing for the WBrC, set a target value for each of the seven attributes the judges will score (for example, prioritising Flavor and Aftertaste), and repeat while measuring the Technical Uniformity of the three cups (ratio, grind size, temperature, turbulence) until the variation is eliminated.
③ A/B comparison of water and recipe. Brew two cups of the same roasted coffee changing only the water composition and score them side by side on the seven attributes; the water and recipe are then fixed on objective evidence such as "B is higher in Sweetness and Overall". The decision is made with a scoresheet, not by feel.
Espresso Tasting — The WBC Protocol
Espresso is concentrated coffee extracted under high pressure. Its evaluation likewise differs from filter — it examines the presence of crema, the tactile sense, and even the harmony with the milk and signature beverages. The World Barista Championship (WBC) is the standard for this.
WBC Espresso is a beverage of approximately 30 mL, poured from one side of a double portafilter in a single continuous extraction. The prescribed figures: water temperature 90.5–96℃, pressure 8.5–9.5 bar, extraction times of the four cups within a 3.0-second spread, and crema covering the entire surface without a break.
| Part | Item | Scale (weighting) |
|---|---|---|
| I. Espresso | Crema (coverage of the entire surface) | Yes/No |
| Accuracy of Taste Descriptors | 0–3 (×2) | |
| Accuracy of Tactile Descriptors | 0–3 (×2) | |
| Taste Experience | 0–6 (×4) | |
| Tactile Experience | 0–6 (×2) | |
| II. Milk | Visual Appeal | 0–3 |
| Accuracy of Taste Descriptors | 0–3 (×2) | |
| Taste Experience | 0–6 (×4) | |
| III. Signature | Accuracy of Taste Descriptors | 0–3 (×2) |
| Well explained & prepared | 0–6 (×2) | |
| Taste Experience | 0–6 (×4) |
In espresso, lipids, fine particles and dissolved solids are concentrated, so the tactile sense (weight, texture, creaminess) is an important element of the experience. The WBC assesses this separately as Tactile Experience. It is an axis that filter does not have.
The order of adjustment after tasting (in-house teaching notes): ① Brew Ratio (overall nuance) → ② Grind Size (detail) → ③ other variables. Aiming for 3.0+ on each WBC Sensory item, keep what is good and, where it is not, adjust starting from the largest lever.
"Espresso must always be 1:2 in 25–30 seconds" — this is a genuinely widespread misconception. 1:2 in 25–30 seconds is only a good starting point in the industry, not an absolute rule; depending on the coffee, the degree of roast and the purpose, a ristretto (1:1.5) or a lungo (1:3) can equally be the right answer. Even the official WBC regulations do not prescribe a ratio or a time (they specify only approximately 30mL and a 3.0-second spread across the four cups).
Alcohol and controlled substances are prohibited in the signature beverage (if detected, that category scores 0). Statements such as "whisky infusions are common in winning signatures" are not true. Only precise knowledge of the regulations avoids a fatal disqualification.
① Designing a new espresso for the menu. Check that the crema covers the whole cup (the definition of the extraction), then, having set the taste descriptors in advance ("blood orange, dark chocolate, silky"), assess separately whether the actual cup matches them (Accuracy) and whether the experience is enjoyable (Experience). Instead of "it simply tastes good or bad", what to fix becomes clear.
② Different extraction targets for different coffees. A light-roast single origin is pulled as a ristretto (1:1.5) to concentrate acidity and sweetness, a medium-roast blend at 1:2 for balance, and a milk base slightly longer to secure contrast against the milk. Rather than being trapped in the 1:2 dogma, the ratio is set to suit the purpose.
③ Diagnosing a tactile problem. If the taste is fine but the cup is "coarse and dry", the tactile axis of the filter is poor — check the uniformity of distribution and tamping (channelling) and check for over-extraction. Just as the WBC assesses Tactile separately, separating taste from tactile sense narrows down the point to fix.
Questions this chapter answers
- How is cupping different from evaluating a brew?
- Cupping fixes grind, water temperature, ratio and time so that several coffees can be compared side by side under standard conditions. Brewed evaluation does the opposite: it judges the result of the recipe that shows that coffee at its best. A coffee that scored well on the cupping table is not guaranteed to be good as a pour-over or as espresso.
- Should the criteria change with processing method?
- The axes stay the same; the expectations differ. From Washed you look for cleanliness and transparent acidity; from Natural, fruit character, sweetness and body. When you cannot tell whether a Natural's fermentative note is a defect or an intention, judge it on whether cleanliness, balance and a pleasant aftertaste hold up.
- How is a cupping score arrived at?
- The SCA method scores individual attributes — fragrance/aroma, flavour, aftertaste, acidity, body, balance, uniformity, clean cup, sweetness, overall — and subtracts for defects. The newer CVA separates descriptive assessment from preference, so that what you perceive and how much you like it are recorded independently.
Figures and regulations are verified against a source hierarchy: Tier 1 in-house knowledge base → Tier 2 SCA·WBC and peer-reviewed literature → Tier 3 web. Anything not adequately supported is marked [to verify] in the text.