Planning for Roasting: Sample Roasting vs. Production Coffee Roasting
Two different skills that should be mastered separately.
When I teach coffee professionals how to cup coffee for different purposes, I tend to say that you can approach coffee from so many angles that you actually need to choose your goal before tasting. You need to know what you are looking for. Otherwise, you are lost. You are there at the cupping table, bombarded with sensory stimuli, and you forget why you were there in the first place.
I can say the same about roasting—both production roasting and sample roasting. The whole process requires so much of your attention and your presence that if you don’t know beforehand what you are aiming for, I can guarantee you will get lost. By that, I mean that after roasting, you’ll feel as if you are emerging from a black hole of time and space: hours passed, you are tired, and you don’t remember anything that happened.
This is why planning in roasting is crucial. What are you doing with every batch? If you are tweaking something, why? If you are repeating a roast profile, how are you doing it?
The Different Roads to the Same Goal
Both sample roasting and production roasting need planning—but totally different kinds. Even though they may have the same purpose (to showcase the coffee at its best for a certain preparation method), the roads to get there are not exactly the same.
Today, I would like to focus on planning sample roasts.
It seems wild to me to approach sample roasting with the mindset of: “Roast all the samples the same so I can see the quality of the beans and not the roast.”
After spending so much time tweaking production roast profiles to get the best out of the coffee, it feels like we just drop the whole idea of using the roast to modulate flavor when we switch to samples. But the fact remains: coffees continue to be extremely different and need a different approach. Applying the same profile to all of them means you will inevitably kill some. Even if you are trying to avoid “roast influence,” the roasting is still there. Roasting itself is modulating and shaping the flavor, whether you do it consciously or not.
If your buying decisions depend on it, a lack of careful sample roasting means you can miss out on some very beautiful coffees. But if you are giving feedback to producers or importers, you have even more responsibility. You must have all the attention in the world to make sure you did your job well.
Production Roasting: The Volume Advantage
In production roasting, our biggest challenge is consistency. We are trying to keep the flavor (not the curve!) consistent while making very educated guesses when tweaking profiles. Every mistake comes in volume. In the best-case scenario, you get a cup that is not at its best; in the worst-case scenario, the coffee is undrinkable but has to be sold anyway.
The good thing about production is also the challenging part: it comes in volume. This means you can make small mistakes or experiments that can be dissolved or corrected. If one thing goes wrong, it’s not the end of the world.
Sample Roasting: One Shot
Sample roasting is very different. In most cases, you have just one attempt. There is a lot of pressure. You have one shot to get the roast right, or at least acceptable.
So, we’d better have a plan.
Sometimes the plan can actually sound like: “I will do things exactly the same for every single batch, see how the coffee behaves, and how it turns out in the cup.”
This is the approach of consistency—but consistency from only one point of view. Consistency here is consistency in application of heat, from one coffee to another (and other parameters, if your sample roasters allows you to change them, like drum speed, fan speed, drum temperature etc etc).
When does this work? It works when the coffees are very similar in processing, size, variety, density, and moisture percentage. When roasting similar coffees, you benefit from not making drastic changes.
Will it guarantee the best possible results? Probably no. But if you have a set of coffees processed the same way from the same region, you won’t be wrong going for that.
Strategies for Different Samples
What if the samples are totally different? What if you have 6 samples from different countries, varieties, and processes? Choosing one profile that fits them all will not work—especially for very dense beans versus soft beans.
Here are two strategies you can apply:
Mimic the Production System: If you are looking for a substitute for a coffee you already have in production, try to mimic your production roaster’s system and transfer that logic to the sample roaster. It’s a lot of work by itself and requires some investigation, but it allows you to understand very quickly if a coffee is worth buying for your specific setup.
Focus on Time and Development: If you don't know what the coffees are good for yet, think Roast Time first and Development second. Try to keep the total time the same and arrive at the first crack with similar momentum. This will give you the most information for future roasts.
The "Learning Curve" Coffees
With some coffees, you will simply get it wrong many times before you get something drinkable.
If you have never roasted Robusta, Conilon, or wet-hulled Indonesia, it will be a learning curve. Nothing helps but trying different things and seeing what works.
Ethiopians also tend to fall in the in-between category, as every Ethiopian coffee is very different, and requires a unique approach. Sometimes things you used to do before work, and sometimes they don’t.
Cupping as an Investigative Tool
Cupping samples is an art in itself (this topic deserves an article on its own). You have to find out what is intrinsic to the bean and what can be improved with roasting.
What does the coffee have that was not developed?
What does it not have (and no roast will ever fix)?
What was done well and can be repeated in production?
When you cup and roast intentionally and with people better than you, you start to see patterns.
