A number on a spec sheet is doing a lot of quiet work. When an offer list says 86.5, that figure travels through your buying decision, your green price, your roast development, and eventually the story on your bag. Yet the score is often the least understood part of the transaction—treated as a grade rather than what it actually is: one trained palate's structured snapshot of a coffee on one particular day.
If you buy green for a living, you already cup. This is less about how to score and more about how to read a score someone else assigned, so it sharpens your sourcing instead of misleading it.
Where the 100-point scale comes from
The Specialty Coffee Association protocol didn't invent taste; it standardized the conversation. A licensed Q grader evaluates a coffee across ten attributes—fragrance/aroma, flavor, aftertaste, acidity, body, balance, uniformity, clean cup, sweetness, and an overall assessment—each scored on a 6-to-10 range in quarter-point steps.
Defects are subtracted. Add the attributes, apply the penalties, and you land on a final figure out of 100. In practice, anything genuinely commercial-and-below rarely gets cupped on this sheet at all, so the working range for specialty sits roughly between 80 and 90. Coffees above 90 exist, but they're rare enough that most roasters will go years without buying one.
The important thing: the scale is not linear in difficulty. Moving from 82 to 84 is common. Moving from 87 to 89 is a different universe of selection, processing control, and luck.
What the bands actually taste like
Numbers mean more when you anchor them to the cup in front of you.
80–83 is clean, correct specialty coffee. No defects, pleasant, but the flavors are general—"chocolate," "nutty," "citrus" without a specific citrus. This is the backbone of most espresso blends and grocery-shelf specialty. Reliable, unremarkable, and often exactly what a blend needs.
84–86 is where character shows up. You start naming things precisely: red apple rather than "fruit," jasmine rather than "floral." Acidity becomes structured instead of just present. An 86 has a clear point of view and a finish that lingers pleasantly. For most roasteries, this band is the commercial sweet spot—distinctive enough to sell as single origin, priced within reach, and forgiving enough to roast at volume.
87–89 is where separation is dramatic and obvious even to a casual drinker. Layered aromatics, sweetness that reads almost sugary, a long clean aftertaste, and balance that holds as the cup cools. These lots command real premiums and reward careful, lighter roasting that protects what you paid for.
90+ is the competition and auction tier. Extraordinary, scarce, and frankly hard to justify unless you're building a specific reputation around it.
So when you ask "what does an 86 taste like?"—it tastes like a coffee that knows what it is. Specific, structured, and memorable, without the fragility or price of the very top.
Why the score is a starting point, not a verdict
A cupping score is accurate and incomplete at the same time. Treat it as one input.
First, scores drift with context. The same coffee can move a point or more depending on roast degree, water, days off roast, altitude of the lab, and the calibration of the table. A score with no accompanying roast date, protocol note, or grader identity is a number without a chain of custody.
Second, the total hides the shape. Two coffees can both score 85 while tasting nothing alike—one carrying a bright, acidity-forward profile, the other soft and heavy-bodied. The attribute breakdown tells you far more about fit than the sum. A roaster building a bright filter program and a roaster building a chocolatey espresso base should read the same 85 completely differently.
Third, a score describes potential under ideal conditions, not what your roast will do to it. High acidity and delicate florals can be spectacular in a competent light roast and completely lost in a rushed dark one.
How to use scores in real sourcing decisions
Read the whole scoresheet, not the total. The individual attribute values tell you why a coffee earned its number, which is the part you can actually act on. High flavor and aftertaste with modest body points to a coffee that will shine as a filter single origin. Strong body and balance with gentler acidity suggests an espresso component.
Cup it yourself, blind, against your own calibration. A supplier's 87 and your 87 need to mean the same thing before you can trust their offer list at scale. Build that alignment on a few lots and future buying gets faster.
Anchor the score to your price ceiling. Each point up the specialty scale costs disproportionately more, and your customers may not perceive the gap between an excellent 86 and a very good 85. Buy the point you can actually sell.
Match the profile to a specific product before you commit volume. The question is never "is this a good score." It's "is this the right coffee for the shelf, the espresso hopper, or the seasonal filter I need to fill."
Finally, weigh consistency over a single peak. A lot that cups 85 reliably across multiple samples and holds through the season is worth more to a production roastery than one 88 you can never replicate.
Scores are a shared language, not the whole conversation
The value of the SCA scale isn't precision to the decimal—it's that a grader in the Western Ghats and a buyer in Portland or Melbourne can describe the same cup and broadly agree. It compresses a complex sensory experience into something you can put on a purchase order and defend to your team.
Our current-season green lots from three partner estates around Chikmagalur are cupped in the specialty range, and we send full scoresheets and samples so you can calibrate against your own table rather than take our word for it—which, honestly, is how it should work.
Use the score to open the conversation. Then cup, ask about the attribute breakdown, request the roast date and protocol, and match the profile to a real slot in your lineup. The number gets you to the table. Your palate, and your customers', decides the rest.