Skip to content
How to Compare Two Coffees Side by Side: A Tasting Framework
Processing & Flavors

How to Compare Two Coffees Side by Side: A Tasting Framework

Two cups tasted together tell you more than either cup alone, but only if the pair differs on one thing. This guide covers how to pick that pair from Indian bags, brew both the same way, keep the second cup from losing by default, and turn what you notice into a buying filter.

ICB
7 min read
Difficulty · intermediate

Say you have a washed and a natural from Chikmagalur. Drink one on Monday and the other on Thursday, and the comparison happens in memory, which is unreliable for something like acidity or body. Drink both within the same ten minutes and each cup becomes the reference for the other.

A side-by-side tasting works like a small experiment, and what it can show depends on what the two coffees have in common. In ICB's catalog, roast level shifts the flavour notes on a bag roughly three times as far as processing does, so a pair that differs on both mostly measures the roast. Building Your Personal Coffee Taste Profile introduces the two-cup exercise. This guide sets out the full method: which pair to buy, how to brew and taste it, and what to write down.

Why two cups work better than one

Perception is relative. "How bright is this coffee?" needs a calibrated memory to answer. "Which of these two is brighter?" does not. Most people can judge direction (more or less acidity, heavier or lighter body, a longer or shorter finish) well before they can name descriptors with any confidence.

That matters because the notes printed on a bag come from cupping panels working under controlled conditions. A home brew usually shows one or two dominant notes, not the full list (see how to read coffee flavor notes). Comparing gets around the vocabulary problem. You can notice that cup A is sharper and lighter than cup B without finding "pomegranate" in either.

The Specialty Coffee Association's Coffee Value Assessment, adopted as its cupping standard in 2024, splits evaluation into two parts: a descriptive assessment that records what is in the cup and how intense it is, and a separate affective assessment of how much the taster likes it. The home method below follows the same order: describe the two cups first, decide which you prefer afterwards.

Step one: choose a pair that changes one thing

Four kinds of pair cover most of what a home taster wants to learn:

  • A process pair: same estate, roaster and roast level, one washed and one natural. This isolates processing.
  • A roast pair: the same coffee from one roaster at two roast levels.
  • A roaster pair: the same estate, process and roast level from two different roasters, which isolates roasting style.
  • An origin pair: the same process and roast level from two estates or regions. This one is less clean, since altitude, variety and often the roaster change along with the origin.

A process or roast pair from a single roaster makes a good first session, because the roaster's style is also held constant.

The roast level is where most pairs go wrong. Among in-stock washed coffees on ICB with flavour notes listed (November 2026), fresh-fruit notes such as citrus, berry or stone fruit appear on 88% of light roasts and 24% of dark roasts. At a fixed medium roast, washed and natural coffees differ by 22 points on the same measure. At light roast they barely differ at all (88% versus 89%). The process difference there shows up in dried-fruit and ferment-type notes and in body, not in whether the coffee reads as fruity. That makes a light-roast process pair the harder test, better kept for a second or third session.

Share of in-stock coffees with at least one fresh-fruit note, by process and roast (ICB, Nov 2026)

Roast levelWashedNatural
Light88% (n=56)89% (n=54)
Light-medium87% (n=47)93% (n=44)
Medium61% (n=134)83% (n=63)
Medium-dark47% (n=85)69% (n=16)
Dark24% (n=45)too few to report

Taste a washed light roast against a natural dark roast and most of what you notice will be the roast. The processing difference is still there, but it is much smaller. For a buyer, that makes the roast level the first label to match, ahead of process or origin.

The mismatch runs in one direction. In 11 of those 13 cases the washed coffee is the darker roast, and across the catalog 53% of naturals are roasted light or light-medium against 27% of washed coffees. Indian roasters tend to roast naturals light and take washed lots further, so a matched pair takes some looking.

Step two: brew both the same way

Keep everything the same except the coffee: dose, ratio, grind setting, water from the same jug, water temperature (just off the boil, around 93–96°C, for most filter methods), brew method, cup type, and timing. Brew the two back to back, within a few minutes of each other. Use the method you actually drink with, whether a pour-over, a French press or a South Indian filter. The result is more useful when it reflects how you normally brew.

Grinders retain a few grams of whatever went through last, so run a small amount of the second coffee through before grinding its dose, or grind the gentler coffee first. Taste both black. Milk compresses differences in acidity and fruit. If you normally drink coffee with milk, a second session with identical milk in both cups is a valid test of its own, but it answers a different question.

Cupping bowls, where coarse grounds steep directly in the cup, are what professionals use because they take brewing technique out of the comparison. They are an option, not a requirement.

Reference doses for a two-cup session

MethodGrindRatioTimeNotes
Pour-overMedium-fine12 g : 200 ml (1:16–17)~3 minnotes: Small doses so both cups are drunk warm
French pressCoarse15 g : 250 ml (1:16)4 minnotes: Pour both presses within a minute of each other
Cupping bowlCoarse8 g : 150 ml4 min steep, then skimnotes: Taste with a spoon once drinkable

Step three: taste across, not down

If you finish cup A before starting cup B, you are back to comparing from memory. Move between the cups instead, and look at one attribute at a time.

  1. 01

    Label the cups

    mark A and B on the underside, or have someone else pour so you taste blind.

  2. 02

    Smell both first

    note which aroma is stronger and roughly which family it sits in: fruit, chocolate, nut, floral, spice.

  3. 03

    First pass, hot

    sip A, then B, then A again, judging acidity first, then sweetness, then body, then finish.

  4. 04

    Rinse and pause

    plain water and a few seconds between coffees.

  5. 05

    Second pass, warm

    about five to ten minutes later, reverse the order and start with B.

  6. 06

    Third pass, near room temperature

    repeat, and note what changed between passes.

  7. 07

    Read the bags last

    compare your notes with the roaster's only after you have written your own.

The three temperature passes come from cupping practice. The SCI cupping protocol, for example, starts tasting around ten minutes after brewing and continues as the coffee cools. Coffee professionals commonly report sweetness and acidity reading differently as a cup cools, though studies disagree on how large the effect is. The passes are a habit worth keeping rather than a rule backed by settled science. A coffee that tastes sharp when hot often reads differently at room temperature, so the first sip is a poor basis for a verdict.

Keeping the second cup from losing by default

Four common sensory errors can skew a comparison.

The first is adaptation: sensitivity drops with continued exposure. In a sensory-lab test run by the green-coffee importer Cafe Imports, a moderately sweet sample tasted after a much sweeter one averaged 1.11 on a 1–3 sweetness scale, against 1.56 when the same sample sat elsewhere in the series. Contrast is related: a delicate washed coffee can seem thin straight after an intense natural. Order effects mean that a coffee always tasted second is always at a disadvantage. And expectation does the rest. Read "strawberry" on the bag and you will go looking for strawberry.

The fixes are simple. Taste the gentler coffee first on the first pass, then reverse the order. Rinse and pause between cups rather than pushing through. Write your notes before reading the bag, and taste blind whenever someone else can pour.

Step four: record direction, then preference

A comparison sheet needs one line per attribute, each marked A, B or no difference:

  • Aroma: which is stronger, and in what family
  • Acidity: which is brighter, and whether it reads as citrus, other fruit, or just sharp
  • Sweetness: which is sweeter, and what kind (cane jaggery rather than brown sugar, say)
  • Body: which feels heavier or more syrupy
  • Finish: which lingers longer, and whether it ends clean or dry

Indian reference points help here. "Mosambi" is more specific than "citrus", and closer to what many Indian tasters actually recognise.

Then add one separate line: which would you buy again, and why. Keeping preference off the description lines stops a liking for one coffee from colouring how you describe the other. It is the same reason the CVA keeps the two assessments apart. Note the roaster, estate, process, roast level and roast date next to the sheet, so the result can be traced to the variable you meant to test.

Pairs you can buy from the ICB catalog

ICB lists 11 roaster–estate combinations (November 2026) that sell a washed and a natural from the same estate at the same roast level. That is 11 out of nearly 1,600 coffees in stock. Few roasters sell a clean process pair side by side, and most of the time the buyer has to put one together. The flavour notes below are the roasters' own, not ICB tasting notes.

For a process pair, Broot Coffee sells two medium roasts from Ratnagiri Estate: Washed Cut (cacao nibs, citrus, dark chocolate, nuts, toast) and Natural Cut (fruity, jam). Hill Tiller Coffee Roasters offers two medium roasts from Stanmore Estate: Marmalade, a washed (citrus, jam), and Jacaranda Tales, a natural (berry, raspberry, toffee, yogurt). Blue Tokai's Kerehaklu Estate light roasts form a washed–natural pair too, but the natural is a fermentation-assisted "Blossom Culture" lot, so the pair tests more than drying method.

For a roast pair, Curious Life Coffee sells Gundikhan Estate washed at two roasts: Vibrant, light-medium (orange, pomegranate), and Bold, medium-dark (black tea, caramel, pomegranate, spice). Pomegranate appears on both bags, which gives you one note to track across the two roasts.

Roaster pairs are the easiest to find. Kerehaklu Estate washed light roasts are in stock from seven roasters, and Ratnagiri Estate coffees from 31. The roaster effect covers what changes between them.

Turning the result into the next bag

One session answers one question. If the washed cup won the process pair, keep the roast level and browse washed coffees at that roast; if the natural won, do the same with natural coffees. If the roast pair settled it, roast level is the strongest filter you have. Change one variable per session. Three or four sessions in, the answers usually start to agree with each other, and the sheets become a record of what you actually like.

References

  1. Cafe Imports — Common Sensory Errors: Adaptation
  2. Specialty Coffee Association — The Coffee Value Assessment
  3. Royal Coffee — How to use the new SCA cupping form
  4. Specialty Coffee Institute — Cupping Protocol v3 (2021)
  5. ICB catalog data — in-stock coffees with flavour notes, snapshot of 8 November 2026

Continue from the field guide

Explore coffees by process and flavour

Filter the catalogue by process and tasting notes.

Browse on ICB →

Enjoyed this article?

Share it with your coffee-loving friends.

About the author

ICB

ICB is a community-driven discovery platform for Indian specialty coffee. Our content draws from 800+ community ratings and a catalog built with input from roasters, enthusiasts, and industry professionals across India.

Frequently Asked Questions

Start with two. Add a third only once you can reliably tell the first two apart. Most home-tasting guides put the useful range at two to four coffees.

No. Brewing both coffees with the method you use every day works, provided dose, grind, water and timing match. Cupping removes brewing technique as a variable, which makes it an option rather than a requirement.

Taste them black first, because milk compresses differences in acidity and fruit. If you usually drink coffee with milk, a second session with the same milk in both cups is a valid separate test.

Usually adaptation and contrast. Sensitivity drops with exposure, and a moderate cup reads weaker after an intense one. Reverse the order on the second pass and rinse with water between sips.

That pair changes the brew method, and usually the roast and chicory content as well, so it shows which way of drinking coffee you prefer rather than which coffee. For a coffee-against-coffee comparison, brew both in the same device.