The SCA Cupping Score Explained: What 84 Points Actually Means
Farms quote scores constantly and almost nobody explains them. The scale is narrower, stranger and more consequential than it looks from the outside.
On this page
Stand in a Colombian farm shop long enough and a number will come up. This lot scored 86. That one scored 84.5. The producer will say it the way a winemaker says a vintage, and most visitors nod without any idea what the number is measuring.
It is worth understanding, because the score is not decoration. It is the mechanism that decides what a farm gets paid, and the difference between two of these numbers can be the difference between a lot going into a commodity container and a lot going to a named buyer at several times the price.
The scale is not what it looks like
It is called a 100-point scale, and that framing misleads almost everyone.
In practice, coffee is evaluated across ten attributes, most of which are scored individually from six to ten. A coffee that is simply drinkable and free of obvious fault lands in the low seventies. Anything genuinely bad is not scored at all — it is rejected as defective rather than given a number.
The functional consequence: the real working range is roughly 50 to 90, and the range that matters commercially is far narrower than that. The action happens between about 80 and 90, which is a ten-point window in which the entire specialty market operates.
This is why quarter-point increments exist. When your whole market sits inside ten points, half a point is a meaningful commercial distinction.
What gets scored
The classic form breaks the cup into ten components. Six of them are scored on judgement:
- Fragrance and aroma — the dry grounds and the moment the crust breaks.
- Flavour — the central impression, the largest single contributor.
- Aftertaste — length and quality of the finish.
- Acidity — quality of the brightness, not quantity of it. A coffee is not marked down for being gently acidic; it is marked down for having harsh or dull acidity.
- Body — again quality rather than quantity. A light-bodied coffee can score well if the texture is fine.
- Balance — whether the components hold together.
Three more are effectively pass/fail across the set of cups, awarded per cup rather than judged on a spectrum:
- Uniformity — do all cups of the sample taste the same?
- Clean cup — is any cup carrying a taint?
- Sweetness — is ripeness present in every cup?
And one is the cupper's overall judgement — the space where a coffee that is technically sound but unremarkable is separated from one that is genuinely exciting.
One bad cherry in a lot will not appear in every bowl. Brewing five cups of the same sample is how a defect that affects a small percentage of the coffee gets caught. A lot can lose several points on uniformity alone while every individual attribute looks fine.
Defects are subtracted, and they hurt
After the attributes are scored, defects come off the total.
The scale distinguishes between a taint — an off-note that is noticeable but not ruinous — and a fault, which makes the cup genuinely unpleasant. Faults are penalised several times more heavily than taints, and both are multiplied by the number of cups affected.
The arithmetic is unforgiving by design. A coffee scoring in the high eighties on its attributes can be dragged below the specialty threshold by faults in two of five cups. This is the structural reason producers are so preoccupied with consistency rather than peak quality: the scale punishes variation harder than it rewards brilliance.
The 80-point line
Eighty points is the threshold at which coffee is formally classified as specialty. It is the single most consequential number in the business.
Below it, coffee is a commodity. It is priced against a global reference, it is blended, and the farm that grew it is anonymous by the time anyone drinks it.
Above it, coffee can be sold on its own identity — farm, variety, altitude, process, producer — and the price is negotiated on merit rather than set by a market that has never tasted it.
Rough working bands, with the caveat that everyone applies them slightly differently:
| Score | Classification | What it means in practice |
|---|---|---|
| Below 80 | Commodity | Blended, priced against the market, origin anonymous |
| 80–84 | Specialty | Sound, distinct, sellable as single origin |
| 85–89 | Excellent | Microlot territory, real premiums, named producer |
| 90+ | Exceptional | Rare, competition and auction lots |
Scores above ninety are genuinely uncommon. If you encounter a farm claiming a 92 as its standard offering, ask who cupped it and under what protocol — which brings us to the real weakness of the system.
Who does the scoring
The score is only as trustworthy as the person who assigned it, and there is no universal registry policing the number on a bag.
The professional credential in this space is the Q Grader certification, which involves a demanding multi-day examination of sensory ability and calibration. A score from a certified grader carries weight. A score assigned by a producer cupping their own coffee alone is a sincere opinion, and producers are not neutral about their own lots.
Reputable operations will tell you which it is without being pushed. When a farm quotes a number, the useful follow-up is simply: who cupped it? It is not a rude question in this industry, and the willingness to answer plainly is informative in itself.
It is also worth knowing that the evaluation framework itself is in transition. The trade body behind the classic form has been rolling out a revised assessment system in recent years that separates descriptive analysis from preference-based scoring, partly in response to long-standing criticism that a single number flattens too much. Adoption across the industry is uneven and the older form remains in wide daily use, so you may encounter either. Anything you read about the current state of that transition, including here, is worth checking against a current source.
Manizales sits at the centre of Colombia's coffee research and training infrastructure, which makes the surrounding area one of the better places to find tours led by people with formal sensory credentials rather than general guides.
What the score does not tell you
Three things, and they matter.
It does not tell you whether you will like it. The scale measures technical quality against a professional standard. A meticulously clean, delicate washed coffee at 87 and a wild natural at 87 are entirely different drinking experiences. Plenty of people prefer an 84 they love to an 88 they respect.
It is a snapshot. Scores are assigned at a moment, on a specific sample, at a specific roast, by a specific person. Green coffee ages. The lot in the bag is not necessarily in the same condition as the sample that was scored.
It says nothing about how anyone was treated. A high score reflects the cup, not the wage paid to the picker, the environmental practices of the farm, or the share of the final price that stayed in the country. Those are real questions and the cupping form does not ask any of them.
Farms selling on score are usually the ones selling direct, and direct-selling farms are frequently the small, rurally sited ones rather than the big roadside operations. A car widens that field considerably. Compare rental cars.
Using the number well
The score is a useful filter and a terrible verdict. Treated as a rough gate — is this coffee in the specialty range or not — it works. Treated as a ranking, it produces the slightly absurd situation of people paying a premium for a coffee half a point higher than one they would enjoy more.
On a farm visit, the better use of the number is as a conversation opener. Ask what the farm's lots typically score, what their best one did, and what they changed to get there. The answer to that last question is where the actual knowledge sits, and it is almost always more interesting than the number that prompted it.