The Impact of Roasting on Coffee’s Flavor

By Rob Hoos

In life, we say that it isn’t so much the destination, but the journey itself that is important. With coffee, it is no different. One can arrive at the same end color—of both whole bean and ground coffee—with similar objective measurements (bean density, moisture content, solubility, etc.) and have the coffee taste totally different based on that path (roast profile) that the roaster took it to make it to their end goal. Recently, in coffee roasting, the focus has shifted from the previous mentality of end color or temperature being the dominant metric for consistency to recording and comparing copious amounts of profile data in order to ensure consistency of product. With the advent of more widely used data logging software (like Cropster, Artisan, Roast Log, and Typica) people have become—rightfully—more interested in how the coffee went from green to a certain shade of brown, and they are easily able to compare roasts day- to-day and week-to-week to see what is happening differently with their roasting.

Even with better tracking and logging, many questions remain outstanding for both professional roasters as well as people just getting started. “How should I go about experimenting with my coffees? How is my profile affecting the final flavor of my coffees? Are there any do’s and don’ts when it comes to roasting coffee? How do people tend to arrive at the profiles they use?” These are the same questions that drove me to 4-5 years of personal research into understanding coffee roasting, the science that seems to lay behind it, and the way that various people all over the world tend to approach roasting. The outcome of much of the detailed experimentation can be read in depth in “Modulating the Flavor Profile of Coffee: One Roaster’s Manifesto.” However, I would like to handle these questions in reverse order for this purpose:

“How do people tend to arrive at the profiles they use?”

In my experience as a coffee roasting consultant, I’ve had the pleasure and opportunity to work with people using most major roaster manufacturers. One of the interesting takeaways for me has been that, by in large, the way most people roast their coffee is the way that seems most natural on their roaster. Allow me to explain. On the Sivetz roaster, there was essentially one setting, which was ‘go.’ Though some people have rigged theirs to have different controls, most people just used it at flat out 100% heat application until the end of the roast. This means that for light, medium, and dark roasted coffee on the Sivetz, we have a relatively short development time compared to other roasting approaches. When a person who is used to a Sivetz tastes a drum-roasted coffee with an elongated development, they tend to not like it because it isn’t what they are used to (which is fine). But it is an example of machine design influencing preference and style.

Another example is for those who roast on drum roasters with a lot of heavy, thermally- retentive material on them. What I have noticed with many such clients is that they tend to have a fairly standard (if you can call anything standard in the roasting community) approach up toward first crack, and then power down (or burner off) before first crack and have a longer development time with a lower rate of rise than people on a less thermally retentive system may have. These are broad brush-stroke examples here, but what I am trying to convey is that most people’s styles and approaches are shaped by the natural tendencies of the equipment they are using as opposed to being intentionally derived and implemented regardless of machinery type. Is this a problem? Not at all! But, I would encourage people to think about how they would want to roast and implement that by working around the tendencies of their roaster instead of just going with the flow. That being said, if the coffee tastes good, then it’s not wrong!

“Are there, then, any do’s and don’ts to coffee roasting that one should follow?”

The answer is a qualified ‘yes.’ There are certain objective and measurable roasting defects that one can create while roasting such as: scorching, tipping, facing, and baking. These should be avoided entirely by the roaster. However, beyond that there is a large world of preference in which nothing is truly wrong, just different. In many ways, when one observes roasting and the approaches people take, there are a lot of similarities, but also a lot of differences. One of the things that I find incredibly enjoyable about our industry is that there is an amazing diversity in the way that people roast coffee and thus a multitude of possible expressions of the same coffee. That diversity of style and approach should be celebrated.

“How is my profile affecting the final flavor of my coffees?”

These differences in roasting approach and profiling affect the flavor of the coffee immensely. As roasters, we have a lot of latitude to manipulate the flavor profile of the coffee through both slight and profound changes in the roasting profile. Though there isn’t enough space to get into it here, I would propose the following changes to flavor profile based on manipulating roast curve. Maillard reaction affects the weightiness of the coffee and thus the body, as well as the complexity of tones in the medium and low end of the flavor spectrum (think ‘roasted, spices, nutty/cocoa’ range on the new flavor wheel published by the Specialty Coffee Association (SCA) and World Coffee Research (WCR).

Development time affects our perception of the type and intensity of the acidity, as well as the intensity of the sweetness. The depth to which we roast the coffee—the lightness or darkness of the final roast—has some to do with our perception of acidity, but more to do with the balance of sweetness and bitterness that exist within the coffee. There are more potential modulations and effects of these changes, but in general terms this is what I find to be true. What’s alarming is that small differences in the roast of 10-20 seconds can be easily perceptible on the cupping table. This means for us as roasters that the devil is in-fact in the details, and once we settle upon a desired roast profile, the need for careful consistency is crucial.

“How should I go about experimenting with my coffees?”

The best way to start playing with a coffee, in my opinion, is right off the cupping roast of the coffee. Whenever we are sample roasting a coffee, we take detailed notes on how we roasted it so we have a general idea of what our starting point was, a potential explanation of why we’re tasting the coffee the way we are, and the baseline I’m working from when I start experimenting with production roast expressions of the coffee. Then, make changes by 30 seconds or 2-5 degrees at a time until the differences are noticeable on the cupping table.

Also, the coffee roaster ought to have an idea or goal for the purpose of that coffee. This will give them a great starting point to work from. For example, if they were using a coffee for a dark roast, a generic profile or end temperature range likely exists already. Modifying the ‘wheel’ may be necessary, but there is no need to reinvent it every time. As another example, if a roaster is buying a coffee because of its incredible acidity, then they’ll likely want to plan their roast to accentuate that. Once a roaster has a general idea of what they’re doing with a coffee, they should make small, intentional changes to one variable at a time in order to get the coffee’s flavor to move in the right direction. While doing this, it is important to drink the coffee prepared in various ways—by pulling shots, using a batch brewer or pour over, and other brew methods—so that one can understand more than just how it tastes on the cupping table.

In all, I would like to encourage coffee roasters to think critically, document carefully, and enjoy the process of dialing in a coffee. Challenging oneself to try the roasting approaches of others—while remembering the bean-temperature thermocouples will likely not read similarly—and drinking other people’s coffee as often as possible for comparison with help roasters evaluate their own product in a more informed manner. We work in an ever- evolving industry where we are learning daily, and that—for me—is why it will always be fun.

rob hoosRob Hoos is an independent coffee roasting consultant as well as the director of coffee for Nossa Familia Coffee. He is also the author of the book “Modulating the Flavor Profile of Coffee: One Roaster’s Manifesto.” Rob feels lucky because coffee was a passion of his first, and now he is fortunate enough to work at something he loves every day. He is a Specialized Lead Instructor for Roaster Level 1 & 2 Certificate Programs, as well as a member of the RG Education Committee. He also serves as a subject matter expert for the SCA.