What Do We Really Know About Roasting Coffee?
This is about something that has been bothering me for some time. But at the same time, it’s exactly what keeps me innovating, looking for new solutions, and questioning how we do things over and over again.
The question is simple: What do we really know about roasting coffee? How much do we actually know?
I bet you that I can summarize almost our entire industry's assumptions in a couple of paragraphs. Not the complex chemical reactions themselves, but the rigid (and at the same time vague) rules we’ve accepted as quality coffee roasters on how to handle heat and what makes a "good" roast.
Let’s play a quick game. Take a moment to think about what rules you would put on a list of absolute roasting truths, and let’s play Specialty Coffee Roasting Bingo:
The Density Rule: Higher density coffees need more energy; lower density coffees need a soft approach to avoid defects.
The “Terroir” Dictatorship: When you chase a chocolatey, caramelization-forward profile, you are just "hiding" the origin and terroir.
The Development Myth: Dark roasted and well-developed coffees are the exact same thing.
The Bake Roast: A longer roast automatically means a dark, flat, and baked flavor.
The Acidity Dogma: Quality coffee must showcase intense enzymatics and high acidity. You either roast dark and developed, to hide the defects, or you roast light to showcase terroir. There is no in-between.
The Processing Rule: Naturally processed coffees have a higher sugar content, need less heat, and are inherently sweeter in the cup.
The RoR Commandments: The Rate of Rise must always be declining. If it flicks or crashes on your screen, you will automatically taste roasting defects like unwanted bitterness, baked or flat.
The Bean Supreme: A bean's intrinsic flavors (variety, terroir, processing) are far superior to any flavors introduced or produced by the roasting process itself.
The Lighter the Better: Light profiles are the only way to preserve delicate organic acids, fruity notes, and floral aromatics.
The Golden Parameters: First crack, development time, and development percentage are the most important metrics of the roast.
I’m not saying I completely disagree with every single statement on this list. But I definitely cannot say I agree with them completely either. Many of them raise far more questions than they actually answer.
As someone who believes there are many different ways of getting from point A to point B — and as a coffee expert focused on roasting stability, quality control, and sensory reality — I naturally started to challenge them.
I started asking myself: Is this really true? What if we try the exact opposite? What actually happens?
What if we accept that human choice of what to plant and how to process coffee has as much influence on the flavour, or even more, than a concept of “terroir” /or “tasting the land”. What if we are tasting not the land, but the producers' choices?
Have you ever experimented with the degree gain, for example? What will change in the coffee flavour-wise if you gain 6 degrees instead of 9 in the same time? Not what you think will change - but what will really change? And in a different system?
Or, let’s say what if we assume that aromatics are not that important? What if there is a way to roast coffee that you get more aromatics retronasally (tasting the coffee) and not orthonasally (smelling the coffee)? How will it change the experience?
What if there is a way to achieve a roast that is easy to brew and extract and that is not dark?
How to make a highly extractable short and light roast?
The answers I’m getting are surprising. I work across completely different roasting machines and roasting systems, and I can tell you for a fact that there are plenty of ways to get from green coffee to incredible flavor. Some of these ways break every rule on a textbook screen curve, yet they absolutely deliver in the cupping bowl, because it looks like coffee doesn't care much about a pretty line on a monitor. We do.
And honestly? I am basically just starting. There is still so much we don't know.
That is my whole point.
Specialty coffee as an industry is only about 30, maybe 40 years old. It is incredibly young. There is not nearly enough deep research done, not enough controlled experiments conducted, and not enough true understanding—even in the sensory department—of how to properly evaluate a roast and map which flavors come from exactly what variable. Not even all the variables are named and defined.
Combine it with the fact that maybe only now are we starting to get some sort of control with advanced equipment, artificial intelligence, and roasting machines that are becoming slightly less unstable. Don’t get me wrong, we are still dealing with a highly unstable system; the machines are just becoming slightly less unstable.
We are finally getting close to measuring what we want to measure, finally having more and more tools, and we can even choose between which roasting system to use, if we want more radiation, conduction or convection - and why. Only now.
The fact that in an industry this young we have already settled on a rigid idea of what coffee roasting is and how it "needs" to be done is ridiculous. It blows my mind that we behave as if everything is already solved and we’ve found the one correct way to roast. I certainly don't. I have worked in this industry for 14 years now, and I prefer to admit that I know maybe somewhere between 1% and 5%.
That knowledge is highly malleable. I am completely open to new information. I am ready to change my mind tomorrow, if needed, and then change it again next week, if I get undeniable evidence on the cupping table that something works differently than I expected.
I would like to believe that coffee roasting in the upcoming years will become more creative and more free. I genuinely hope that at some point we stop chasing the lines on the screen, and start experimenting more, and truly understand what happens to flavor when we make those green beans brown.
We are just getting started.
