Quick answer: Origin sets a coffee’s flavour potential: variety, altitude, processing and green quality decide which flavours are available. Roasting decides how much of that potential reaches the cup. Research on trained tasting panels found roast colour is the strongest single predictor of flavour, and at the same colour, development time after first crack matters more than time to first crack. Darker or longer roasts add bitterness and reduce acidity, fruitiness and sweetness. Profile by changing one variable at a time, measuring roast colour, and cupping to the SCA protocol.
Roast profiling can feel like guesswork: dozens of variables, strong opinions, and recipes that never transfer from one roaster to another. This guide focuses on what holds up across machines: what origin contributes, what physically happens in a roast, what peer-reviewed sensory research says about roast colour and timing, and a repeatable way to build and check your own profiles.
What Origin Contributes to Flavour
“Origin” is shorthand for several things that travel with a green coffee:
- Variety and genetics shape the aromatic precursors in the bean.
- Altitude and climate affect how slowly cherries ripen and how dense the beans are. Denser beans generally take heat differently from softer, lower-grown coffees.
- Processing (washed, natural or honey) changes sweetness, body and fruit character. See our guide to honey process flavour profiles.
- Green quality. Under the SCA grading protocol, specialty-grade green coffee must have zero Category 1 defects and no more than five Category 2 defects in a 350 g sample. The SCA standard also sets a water activity below 0.70 aw for specialty green coffee.
The table below summarises how coffees from major regions are commonly described in specialty cupping. Treat it as a starting map, not a rule: a single farm, lot or process can sit well outside its region’s stereotype.
| Origin | Common processing | Often described as | Roasting starting point |
|---|---|---|---|
| Ethiopia (washed) | Washed | Floral, citrus, tea-like | Lighter roasts usually preserve delicate aromatics |
| Ethiopia (natural) | Natural | Berry, tropical fruit, winey | Watch for heavy, jammy flavours if development runs long |
| Kenya | Washed | Blackcurrant, bright and juicy acidity | Lighter to medium to keep acidity vivid |
| Colombia | Washed | Caramel, red apple, balanced | Versatile; works for filter and espresso |
| Brazil | Natural or pulped natural | Chocolate, nuts, low acidity, heavy body | Common espresso base; medium roasts are typical |
| Central America | Mostly washed | Cocoa, citrus, honey sweetness | Balanced coffees, useful for learning profile changes |
| Sumatra | Wet-hulled | Earthy, herbal, syrupy body | Often roasted medium to medium-dark |
What Happens to a Bean During Roasting
Green Arabica is roughly 12% water. During roasting it loses between 15% and 22% of its weight, mostly as water, and finishes at about 1–2.5% moisture. Its volume grows by about 50% to almost 100% (Münchow et al., 2020). Most of the flavour is created by non-enzymatic browning, such as the Maillard reaction and caramelisation.
| Phase | What happens | What you control |
|---|---|---|
| Drying (charge to colour change) | Water heats and evaporates; beans turn from green to yellow | Charge temperature, early heat input, airflow |
| Browning (colour change to first crack) | Maillard reactions build colour and aroma; steam pressure rises | Rate of heat input, airflow, drum speed |
| First crack | Steam pressure cracks the beans with an audible pop | Heat reduction timing to avoid a stall or a surge |
| Development (first crack to drop) | Fast aroma formation; colour darkens; acidity falls and bitterness rises | Development time and end colour |
The bean temperature reading at first crack depends on the roaster and probe placement, so figures published elsewhere rarely transfer directly. Log your own first-crack temperature and use it as your reference.
What the Research Says: Colour vs Time
In 2020, Münchow, Alstrup, Steen and Giacalone published eight sensory studies that used trained or semi-trained tasting panels on washed Arabica roasted in Probat and Loring drum roasters. Their main findings are directly useful for profiling:
- Roast colour was the stronger predictor of flavour. Darker roasts were associated with more bitterness and less acidity, fruitiness and sweetness.
- At the same colour, development time had a larger effect than time to first crack. Longer development reduced acidity and fruitiness and increased roasted, bready notes.
- Total roast time on its own was not correlated with any flavour attribute. Where the time is spent matters more than the total.
- Body did not change with either roasting phase, which goes against a popular belief that a longer Maillard phase builds body.
The authors caution that their samples were washed coffees in a mid-range colour band, so results may differ for naturals or very light and very dark roasts. Still, the practical message is clear: measure colour, and treat development time as your main flavour lever.
How to Build a Baseline Roast Profile
- Pick one coffee and one goal. For example, “a filter roast that keeps the citrus”.
- Fix a target colour and measure every batch with a roast colour meter if you can.
- Change one variable per batch. Start with development time, holding charge temperature and airflow constant.
- Log key events: colour change, first crack (time and temperature), and drop. Many roasters also track development time as a share of total time so batches of different lengths can be compared.
- Rest and cup side by side using the protocol below, then keep the version that best meets your goal.
Cupping Your Roasts the SCA Way
Cupping is how you close the loop between the roast log and the cup. The SCA protocol specifies:
| Parameter | SCA protocol |
|---|---|
| Sample roast | 8–12 minutes, cupped 8–24 hours after roasting |
| Cupping roast colour | Agtron Gourmet 63 (ground), ±1 |
| Ratio | 8.25 g whole-bean coffee per 150 ml water |
| Grind | 70–75% of particles pass a US 20-mesh sieve |
| Water | About 93°C; 125–175 ppm total dissolved solids ideal |
| Steep | Crust left 3–5 minutes before breaking |
| Tasting | Begin when the cup cools to about 71°C |
Our guide to developing a specialty coffee palate covers how to train your tasting, and the CQI Q Grader programme is the professional route if you want formal calibration.
Troubleshooting by Taste
| What you taste | Likely roast cause | First thing to try |
|---|---|---|
| Harsh bitterness, ashy finish | Roast too dark or development too long | Drop earlier or lighter; shorten development |
| Flat, bready, dull (the “baked” defect) | Development stretched too long | Shorten development; keep heat input steady into first crack |
| Sharp, sour, grassy or peanut-like | Underdeveloped roast | Extend development slightly or finish a little darker |
| Fruit present but muted | Colour darker than the coffee needs | Aim for a lighter end colour |
Taste problems are not always the roast. Uneven extraction and water chemistry can produce similar defects, so check your brewing with our brew ratio calculator, our explainer on coffee extraction science, and our guide to water chemistry for extraction before changing your profile.
Frequently Asked Questions
Does coffee origin or roast matter more for flavour?
Both, in different ways. Origin (variety, altitude, processing and green quality) sets which flavours are available. Roasting decides how many of them survive. Research by Münchow and colleagues found roast colour was the strongest single predictor of flavour, with darker roasts tasting more bitter and less acidic, fruity and sweet.
What is development time in coffee roasting?
Development time is the time from the start of first crack to the end of the roast. In a 2020 multi-study analysis, development time had a bigger effect on flavour than the time taken to reach first crack, even when final roast colour was held constant.
At what temperature does first crack happen?
The reading depends on your roaster and where the probe sits, so published figures vary. Log the bean-probe temperature at first crack on your own machine and use that as your reference rather than a number from another roaster.
How do you cup coffee to SCA standards?
Use 8.25 g of whole-bean coffee per 150 ml of water, ground slightly coarser than filter so 70–75% passes a 20-mesh sieve. Pour water at about 93°C, let the crust steep 3–5 minutes, break it, and start tasting when the coffee cools to about 71°C.
Keep Learning
- Step-by-step guide: how to cup coffee at home
- Honey process coffee: flavour profiles explained
- Developing a specialty coffee palate
- Coffee extraction science
- Fair trade and single-origin certifications
- Black Rifle Coffee review: roasts, prices and Coffee Club
Sources
- Münchow, M., Alstrup, J., Steen, I. & Giacalone, D. (2020). Roasting Conditions and Coffee Flavor: A Multi-Study Empirical Investigation. Beverages 6(2), 29.
- Specialty Coffee Association. Protocols and Best Practices (revised 2018). Green grading and cupping protocols.
- Specialty Coffee Association. Coffee Standards (archived copy). Water activity standard for green coffee.
Leave a Reply