How We Taste-Test New Coffees (the Cupping Process, Plain Version)
Cupping sounds like something only professionals do, in a lab, with a $2,000 setup.
It isn't.
Cupping is just brewing coffee in a standardized way so you can taste it fairly. Anyone with a kettle, a scale, and a few cups can do it. We do it weekly when we're evaluating new coffees for the catalog.
This is the version we'd run for a friend — what cupping actually is, how we do it, and how you can run a version at home with no specialty gear.
What cupping is (and why we do it)
Cupping is a brewing method designed for comparison, not for daily drinking. Every variable that could change the taste — water temperature, grind size, brew time, water-to-coffee ratio — is held constant. The only thing that varies is the coffee itself.
When you're tasting 8 or 10 coffees at once and trying to decide which ones to stock, this matters. If you brew each one in a different French press or pour-over, the cup difference might be the brew method, not the bean. Cupping removes that variable.
What you taste in a cupping:
- Body (light, medium, full)
- Acidity (bright, balanced, dull)
- Sweetness (sugar-like, fruity, caramel, chocolate)
- Flavor notes (specific: blueberry, cocoa, almond, citrus)
- Finish (clean, lingering, short)
- Balance (do the components work together)
- Defects (anything that tastes off — fermented, musty, sour in a bad way)
What cupping doesn't tell you:
- How the coffee will taste as espresso
- How it will taste as cold brew
- How it will taste with milk
- Whether you will like it for your morning
A coffee that scores well in a cupping can still be wrong for a particular use. That's why cupping is the first step, not the only step.
How we run a cupping session
Here's the process we use when we're evaluating new samples. It's based on the Specialty Coffee Association protocol, simplified for a small operation.
The setup
What you need:
- 8–10 identical cups or bowls (we use 5-oz white porcelain cupping bowls, but any small bowl works)
- A kettle
- A scale (gram-accurate)
- Cupping spoons (or any deep soup spoon)
- Sample coffees (15–20g each)
- A grinder (burr, set to a coarse grind slightly coarser than filter)
- Water at 200°F (93°C) — fresh off the boil, rested 30 seconds
- A timer
- Notepad or scoring sheet
Setup steps:
- Label each cup with the coffee name (use sticky notes on the bottom, or number them and track on paper)
- Grind each coffee separately — 15g of coffee per cup, ground right before brewing
- Place grounds in each cup
- Heat water to 200°F and have it ready
The brewing (4 pours per cup)
Cupping is a sequence of timed steps. Here's the full protocol.
Pour 1 — the bloom (0:00)
- Pour 200°F water onto the grounds until they're fully saturated. Use roughly 30g water per 15g coffee (1:2 ratio).
- Wait 30 seconds. The grounds will bubble up as CO2 escapes — this is normal.
- The bloom tells you about freshness. Fresh coffee blooms aggressively. Old coffee doesn't bloom at all.
Pour 2 — the full brew (0:30)
- Add water to reach the target ratio: roughly 250g water per 15g coffee (1:16.6 ratio).
- Set a timer for 4 minutes.
- Let it sit undisturbed.
Pour 3 — the crust break (4:00)
- At the 4-minute mark, gently break the crust of grounds on top with a spoon. Push the grounds aside while you smell.
- The aroma here is the most information-dense moment in the cupping. You'll catch the dominant flavor notes and any obvious defects.
Pour 4 — the tasting (4:30 onward)
- Skim the foam and floating grounds off the top.
- Use your cupping spoon to scoop a small amount of liquid. Slurp it — yes, slurp — to aerate the coffee across your whole palate.
- Note what you taste.
- Repeat with the same coffee as it cools (5 minutes, 10 minutes, 13 minutes). The flavor profile changes as the coffee cools, and good coffees often reveal more complexity at lower temperatures.
Total time: ~15 minutes for the full sequence across all cups.
What we're scoring
For each coffee, we score across these categories. We use a 10-point scale per category, with notes:
- Fragrance/Aroma (dry grounds + wet grounds after the bloom)
- Flavor (specific notes: fruit, chocolate, nuts, floral, etc.)
- Aftertaste (how the flavor lingers)
- Acidity (brightness, fruit-like sharpness, balance)
- Body (mouthfeel — light tea-like, medium, heavy syrupy)
- Balance (do the components work together or fight each other)
- Overall (the gut-level "would I buy this" reaction)
Specialty-grade coffee scores 80+ on the SCA scale. We're not picky about exact numbers; we are picky about balance and clarity.
What we look for
Three things matter most when we're picking coffees:
1. Distinctiveness. Does this coffee taste like something we don't already have? If it's "fine, balanced, chocolatey" but we already stock a fine, balanced, chocolatey coffee, it doesn't add anything.
2. Clarity. Are the flavor notes identifiable, or is it just "coffee-flavored"? The best coffees have specific notes you can name — dried orange, blackberry, cocoa, almond butter. If we can't pick out specifics, the coffee is probably average.
3. Balance without blandness. A coffee can have big, distinctive notes and still be poorly balanced (the brightness overwhelms the sweetness, the body is too heavy for the acidity). We want notes that are loud but the coffee still feels composed.
What we reject
We pass on coffees that:
- Score below ~83 on the SCA scale (specialty-grade minimum)
- Have visible defects in the cup (fermented, musty, sour in a way that doesn't resolve as it cools)
- Taste like something we already stock without offering a new angle
- Are from origins or processes we want to support but where the execution wasn't there
- Come from producers or importers we can't verify (we need to know who grew it and how it got to us)
That's roughly 60–70% of the samples we receive. The pass rate is high because we'd rather have a small, intentional catalog than a large one with mediocre options.
What cupping does NOT tell us
This is the part most cupping posts skip. Cupping is one tool. Here's what it misses.
It doesn't tell you how a coffee performs with milk. A bright, acidic Yirgacheffe can cup beautifully but turn sour in a latte. A heavy-bodied Sumatra might cup flat but shine in a milk drink. We always brew test candidates as a small batch — usually 100g — in a French press before committing to a stock order.
It doesn't tell you how the coffee ages. Freshly roasted and 14 days post-roast taste different. The cupping happens on day 5–7 post-roast for green samples. We don't always know how a coffee will taste at the 3-week mark when it actually arrives in customer bags. That's why we run small production batches first.
It doesn't tell you how the coffee will perform in espresso. Espresso is a different extraction. We'll usually pull 2–3 shots from a candidate before deciding if it's espresso-worthy. This is how we ended up with the Whiskey Barrel Aged as our espresso pick — it pulled sweet and complex, where most of our other origins pulled flat or harsh under 9 bars.
It doesn't tell you what your customers will like. Cupping reflects our team's preferences. Our customers have different preferences. We try to keep the catalog varied enough that anyone drinking coffee will find at least 3 or 4 they love.
How to cup at home (the simplified version)
You don't need a $200 cupping kit. Here's the 80/20 version for home.
Equipment:
- 4–6 small bowls or cups (white is best — color tells you things)
- A kettle
- A scale (or measuring spoons, less precise but workable)
- Cupping spoons (or any deep spoon)
Process:
- Weigh 15g of coffee per bowl (or use a heaping tablespoon if no scale)
- Coarse-grind each coffee separately
- Pour 200°F water to about 250g per bowl
- Wait 4 minutes
- Break the crust, smell, skim
- Slurp small spoonfuls, note what you taste
- Repeat as it cools
Total time: 15 minutes for the full sequence. Total cost: $0 if you have bowls and a kettle.
What you'll learn: a lot, fast. Cupping 4 coffees side by side makes differences obvious in a way that drinking them on different days never does. The same Colombia will taste different from a Costa Rica, which will taste different from a Brazil. Once you've cupped them side by side, you'll taste those differences in every cup for the rest of your life.
What you won't be able to do: run this with espresso or milk drinks easily. Those need different equipment.
A typical week for us
For context, here's what a coffee-selection week looks like behind the scenes.
Tuesday: New samples arrive from importers. We log them — origin, producer, process, harvest date, price — and assign each one a tracking number.
Wednesday (roast day): We roast small 100g batches of every sample. Roasting small is harder than roasting production (less thermal mass, more variation), but it lets us taste each one before committing to a 5-pound batch.
Thursday or Friday: Cupping session. We taste 6–10 samples, blind if possible (we shuffle the order so we don't bias toward the sample we just talked about). Each sample gets a score sheet and a one-line gut reaction.
Saturday: Brew tests. The 2–3 samples that scored highest go into a French press and a small espresso pull. We taste them as actual brewing methods, not just cupping.
Sunday: The decision. We pick 0–2 to add to the catalog, write up the tasting notes, and figure out the launch roast schedule.
Then: we order 5–15 pounds of the green coffee, depending on how confident we are. Smaller first orders mean smaller risk if a coffee doesn't sell.
This whole process can take a few weeks to a few months from sample to shelf. Most of our catalog started with a 100g sample that scored well.
The honest version
Cupping is a tool, not a gatekeeper. We've passed on samples that ended up doing well for other roasters, and we've stocked samples that scored well but didn't sell. Cupping tells you what we think. It doesn't tell you what you'll think.
If you're curious about how our current catalog would cup — and most people aren't, but if you're reading this you might be — our How We Pick Which Coffees to Stock post walks through the sourcing side. Our How to Taste Coffee Like a Pro post is the consumer version of what we do in the cupping room.
If you want to actually cup our coffees at home, the easiest starter set is to order 3 single origins (we'd suggest Colombia, Costa Rica, and Brazil Santos — three distinct flavor profiles), grind each separately, and run the protocol above. You'll taste the difference between washed and natural processes, between high-altitude brightness and lower-altitude body, between citrus-forward and chocolate-forward profiles — in one sitting.
Cupping isn't magic. It's just standardized brewing with a side-by-side comparison. We do it because the catalog depends on the picking, and the picking depends on tasting the same way every time. You can do it at home with a kettle and a few bowls. The most useful thing it teaches is what specific coffees actually taste like — and once you have that reference, every cup you drink becomes more interesting.
Brew slow. Sip slow. ☕
P.S. The single biggest variable in any cupping is the water. If you brew with hard, chlorinated tap water, you'll taste less difference between coffees because the water is masking the flavor compounds. Our water quality post covers the $30 fix that makes cupping — and every other cup — clearer.