Cupping and tasting

What you will be able to do

• Set up and run a simple cupping at home
• Explain why cupping uses such a plain brewing method
• Use a triangle test to check whether a difference is detectable
• Name three biases that affect tasting and how to reduce each
• Write a tasting note someone else could use

Assumes: Level 10: Sensory foundations.

Why cupping looks the way it does

CoffeeHQ explanation of the mechanism

Cupping is how the trade tastes coffee. Ground coffee is put in a bowl, hot water is poured on, it steeps, and the liquid is tasted from a spoon. There is no filter and no technique to speak of. That plainness is the point. The method is designed to remove the brewer from the comparison, so that differences between bowls are differences between coffees.

It is also easy to repeat exactly, anywhere, with dozens of samples at once. A buyer in one country and a seller in another can cup the same coffee the same way and discuss the same thing. It is a tool for evaluating and comparing; it is not meant to be the best way to drink a coffee.

In more depth: Coffee cupping explained · Cupping coffee

The procedure

From published standards or research

The Specialty Coffee Association's cupping standards specify a ratio of 8.25 grams of coffee to 150 millilitres of water, water at about 93 °C, a medium-coarse grind, and several bowls of each coffee so that a single defective bean shows up as one odd bowl.

The sequence is fixed. Smell the dry grounds. Pour the water and let the coffee steep; a crust of grounds forms on the surface. At about four minutes, break the crust by pushing a spoon through it and smell the aroma released. Skim off the remaining foam and grounds. When the coffee has cooled enough, taste it by slurping from a spoon, which sprays the liquid across the mouth and carries aroma to the nose. Taste again as it cools, because coffees change with temperature and faults often show most clearly when cool.

Scoring has changed. For two decades the association's form produced a single score out of 100, with 80 points as the conventional threshold for 'specialty'. In November 2024 it adopted the Coffee Value Assessment as its standard, replacing the 2004 protocol and form. The new system separates describing a coffee from judging how much one likes it. The 100-point language is still widespread on bags and in the trade.

In more depth: Cupping spoon · Coffee grading and quality

The biases tasting has to control

From published standards or research

Tasters are not instruments. Sensory science describes many ways that judgement is pushed around, and professional practice is largely a set of defences against them.

Expectation: knowing a coffee's price, origin or reputation changes what you perceive. The defence is tasting blind, with samples identified by code. Order and contrast: a coffee tastes different after a very acidic one than after a flat one, and the first sample is judged differently from the last. The defence is to randomise order and re-taste. Suggestion: one person saying 'strawberry' aloud makes the table taste strawberry. The defence is silence until everyone has written their notes. Fatigue: palates dull over a long session. The defence is fewer samples, water and breaks.

The triangle test is the simplest formal check. Three cups, two the same and one different; the taster must pick the odd one. Chance alone gets it right a third of the time, so repeated correct answers are evidence that the difference is real to that taster. It answers one question only — can the difference be detected — and is the right first test for any claim that something 'makes a difference'.

In more depth: Comparing two coffees

Calibration and useful notes

CoffeeHQ explanation of the mechanism

Calibration is the process by which a group of tasters comes to use words and scores the same way, by tasting the same samples and comparing notes. It is why a trained panel can agree, and why an individual's descriptions drift without one. Tasting with other people, even informally, is the single most effective way to improve.

A useful note records, in order: what the coffee was and how it was brewed; the attributes, briefly — acidity, sweetness, body, finish; one or two flavour families; any fault; and whether you liked it, kept clearly separate from the description. 'Bright, light body, citrus and floral, clean finish, liked it' is a better note than a paragraph of fruit names, because someone else can check it.

In more depth: Tasting Lab worksheet

Practise with a tool

Exercise: A first triangle test

See all practical exercises

Work it through: Everyone tasted peach

Four friends cup three coffees together. The most experienced says of the second bowl, 'wow, peach'. All four notes for that coffee say peach. Nobody mentions peach for the same coffee when it is tasted again a week later, blind.

Reasoning it out

1. Four independent tasters rarely choose the same specific fruit.
2. The descriptor was spoken before anyone wrote, by the person whose opinion carried most weight.
3. That is suggestion: the notes record agreement with a person, not four observations.
4. The blind re-taste is the better evidence of what each person perceives.

What to do

Taste in silence and write first; discuss afterwards. The first session produced one data point repeated four times. Run that way, a group is a calibration exercise instead of an echo.

Check yourself

Answer in your head first. Nothing is scored and nothing you open here is recorded.

1. 1. Why does cupping use no filter and no pouring technique?

   Show the answer

   To remove the brewer as a variable, so bowls differ only in the coffee.

   It is a comparison method, not a serving method.

   Back to “Why cupping looks the way it does”

2. 2. By chance alone, how often would you pick the odd cup in a triangle test?

   • One time in two
   • One time in three
   • One time in four
   Show the answer

   One time in three.

   Which is why a single correct answer proves nothing and several in a row is evidence.

   Back to “The biases tasting has to control”

3. 3. Name the bias controlled by tasting in silence.

   Show the answer

   Suggestion — being led by what someone else says.

   Write first, then compare.

   Back to “The biases tasting has to control”

4. 4. Why are several bowls of each coffee prepared?

   Show the answer

   So that a single defective bean shows up as one odd bowl, and consistency can be judged.

   Uniformity between bowls is itself part of the assessment.

   Back to “The procedure”

What a page cannot teach

Cupping scores from a calibrated panel and notes from one person at home are different kinds of evidence. This level teaches the method; it does not calibrate you.

Reference for this level

• Exercise: Labelled, then blind
• Coffee competitions and professional roles
• Coffee sensory and tasting basics
• Flavour notes

Where the facts in this level come from

• SCA cupping standards and the Coffee Value Assessment — Specialty Coffee Association
• Coffee Quality Institute Q Grader / cupping standards — Coffee Quality Institute (CQI)

Sections are labelled with the kind of claim they make. How CoffeeHQ uses sources, and what each one is relied on for, is set out in the source and evidence policy.