What We Do: Roasting

A phrase readily tossed around in the coffee community is “Seed-to-Cup,” but what does it actually mean? In an effort to pull the curtain aside for everyday coffee consumers, we’re introducing a new series called “What We Do.” We hope to give our community an in-depth look into the journey that Blueprint’s coffee takes from the farm to our cafe. For our second post, we’ll explore the fundamentals of roasting and what makes our process special.


In our first “What We Do” post, we covered our sourcing. We focused on the importance of collaborating with farmers to create mutually beneficial relationships, better products, and an overall healthier, more sustainable industry. For Blueprint, roasting is an extension of sourcing. We already see the potential in a coffee when we decide to add it to our roster. So, roasting becomes our method to highlight and represent the farmer’s work by finding the right balance of flavors. We define a good roast as one that showcases the best qualities of a coffee while still leaving room for interpretation and new experiences.



So how does roasting work? Simply put, roasting is the process we use to transform green coffee into the final product our customers drink. During roasting, we manipulate and apply heat to raw coffee seeds in a controlled environment, not dissimilar to cooking or popping popcorn.

The amount of time roasting takes is largely dependent on the machine, batch size, and type of coffee. Most people imagine a longer process than it actually is. A sample roast, which is about a half-pound of coffee, takes 6-9 minutes from start to finish to roast. Our 20-40 pound production roasts take about 11-13 minutes.

Once the roast is complete, the coffee drops onto a cooling tray and takes about 5-7 minutes to cool down. After cooling, we weigh the coffee and record the amount of weight loss that occurred during the roasting process. We look at weight loss as one data point when analyzing how to roast future batches. From there, we hand scoop the coffee onto a scale, pour it into bags, and seal it. Then, we send it out to wholesale partners or put it directly on our shelves.



For both sample and production roasts, coffee goes through three main stages of transformation during roasting: drying, yellowing, and development. During the drying phase, we first apply heat to the raw product, much like searing meat or toasting bread. We add just enough heat to allow the coffee to begin absorbing it. Then, chemical changes start occurring. After this, the coffee goes through the physical process of changing colors known as the yellowing or Maillard phase.

During this second phase, the coffees shifts from pale green to yellow to orange to cinnamon, and then to brown. Once the coffee moves into the brown coloring it will begin to break down and pop open the way that popcorn might. The industry calls this first crack, and it indicates that the last phase, the development phase, has begun. The length of the development phase depends on how light or dark coffee is being roasted.

"If as a company we enjoy a taste experience that is different from our expectation of a region or variety, that tends to excite us and make us want to share that experience with our customers.”

-Andrew Timko

Andrew Timko sorts through coffee that has just left the roaster and is cooling.

When darkly roasted, the processes' second crack will occur in which the popping of the first crack happens all over again. In the development phase, as little as 10-15 seconds can determine if a coffee is lighter or more darkly roasted. We roast our coffees on the lighter side, so we make sure the beans aren’t in the roaster long enough for the second crack to happen.



The reason we don’t venture into the second crack during the development phase nods back to our roasting philosophy. We want to achieve a coffee’s maximum potential. For us, that means capturing the flavor and experience that is already stored inside the coffee bean. By roasting a coffee on the lighter side, we avoid leaving our coffee in the roaster long enough for bitter and smoky flavors to develop. We believe these characteristics overpower the more unique qualities of our coffees.



In the process of coaxing out the nuanced flavors held within a coffee, we also work towards creating an overall composition that balances the body, brightness, and sweetness of a coffee. The body alludes to the viscosity of the coffee. With beverages, viscosity explains the mouthfeel difference between drinking a cup of tea and a cup of chocolate milk. We want our coffees full-bodied enough to hold room for acidity, sweetness, and more defined flavors without being so viscous that the coffee tastes muddy.

When we refer to the brightness of a coffee, we’re really talking about its acidity, much like the bright acidity of citrus. Bringing out brightness in a coffee can help sharpen and highlight its inherent flavors. Sweetness, the third aspect we focus on in our flavor composition, is more straightforward. It’s the sweet flavors like chocolate or brown sugar that can be found in a cup.

When taking these three aspects into consideration, we always try to make sure that none of them become one-dimensional or too dominant in the final product. Our coffees strike a balance of body, brightness, and sweetness. Even if we highlight one quality slightly more than the others, we want to maintain a drinkable and desirable cup.



The manipulation of a coffee’s flavor happens during the three phases of roasting. The amount of heat applied and time spent during each phase can drastically alter the coffee’s overall composition. The more time the coffee spends in heat, the more it breaks down and becomes vulnerable to changes. In this process, a coffee's overall profile can change from a more acidic one to a sweeter one in a matter of seconds.

We seek to roast coffees that are unique and diverse.

“We like to buy a higher quality product that might have more interesting, nuanced characteristics that we would like to highlight rather than making a general, uniform flavor.”

-Mazi Razani

For example, if we want to produce a coffee that is brighter or more delicate, we might give it less development time. Then, the subtle qualities don’t get destroyed by too much heat. On the other hand, if we’re trying to pull chocolatey flavors out of coffee, the development period might be longer. This allows sweetness to develop through caramelization. Generally, we experiment more with the length of the first two phases of roasting to manipulate nuanced flavors. We tend to keep the development phase length fairly consistent from coffee to coffee.



Being able to control and carefully monitor heat throughout the roasting process is essential to detailed flavor and composition development. Our knowledge, skillsets, and decisions are crucial to the roasting process, but they would be rendered useless without the aid of our exceptional equipment.

Our roaster is a 1937 23-kilo Gothot, and is a workhorse for Blueprint Coffee.

Gothot is a German manufacturer that was purchased by Probat in the mid 1970’s. Before Probat bought the company, Gothot patented a unique tangential heat application design in 1969. The design allowed air to be directed through the side to help transfer heat more evenly throughout the drum while the coffee was moving through the drum. We found our roaster through a company that restores used equipment and our timing was perfect. In 2012, Kevin Reddy and Andrew Timko went to visit the facility to test it out and fell in love.

One of the most unique things about our roaster is its thermodynamics. We have more control over heat and variable controls than most modern-made roasters. Many modern roasters are designed to have one heat source, a flame underneath the roasting drum. This places the coffee directly on top of the heat. Our Gothot differs in that the heat source comes from the side where a fan pulls hot air into the roaster. This allows for more hot air control and an even heat application to the beans. Essentially, most modern roasters are built like conduction ovens. Our Gothot uses more convection, which allows air circulation to create more even roasting.

“I tasted it, and it tasted like a completely different coffee and was just so much better than what I was tasting the week prior. The small things make such a large difference and sometimes wish you could be Jimmy Neutron and figure out why and what is chemically changing in these coffees that lets you achieve what you want from them. As time goes on, I’ve realized that you limit yourself by locking into a certain way of working all the time, and experimentation often can lead to things you would not have expected.”

-Mazi Razani

Mazi Razani checks the roast data in real time during coffee roasting using Cropster.
Mazi tastes the results of his roasts using the cupping method.


We have three temperature probes set up in the roaster that let us monitor the entire roaster’s environment. One tracks the temperature of the heat source, which is the hot air going into the roaster. Another is on the exhaust of the roaster, which lets us monitor the hot air going up and out of the roaster. The last one tracks bean temperature and sits towards the top of the roaster drum, allowing it to measure the environmental temperature inside the drum. The collected probes help us observe the absorption of the heat during the different phases of the roast.

During the roast, there are constantly changing variables like batch size; retained drum heat, convective heat transfer, and barometric pressure that can all create variances in the roasting process. We use a software from Cropster called Roasting Intelligence. It maps the temperatures collected from our probes in real time, giving us the ability to break down the data monitored during a roast. We use that data to establish more controls. If we evaluate a coffee as we roast, make a change to the next roast, and document what effect that change had, we are essentially able to investigate how the smallest of details can change the overall profile of a coffee.


Collection, documentation, and analysis of data gives us more control over the roasting process. Still, it’s important to remember that we are working with an organic product. It changes from seed to seed, day to day, and region to region. Roasting coffee is like shooting at a constantly moving target. So, trial and error is always going to be part of the process. The physical characteristics of a bean, like its size, density, altitude, and the region, alter our approach to a roast. In addition, so do details as seemingly insignificant as the temperature or humidity outside on a roast day.

Sometimes, we have a more set approach to a coffee. This depends on the region of the world it’s from or the way it was processed. For example, most of the time we’ll roast an Ethiopian coffee completely different than a Colombian coffee. Or, we'll roast a washed Ethiopian coffee differently than a natural. Even then, the more we hypothesize, test, and experiment, the more we see the benefits of frequently trying new approaches. Even if two coffees are produced in the same region, a farmer's practices might not be the same as their neighbors.

Andrew Timko tastes through sample roasts from an importer.

“Our approach to working with a new coffee comes from our philosophy. We try to remain open to the potential of a coffee and taste the initial roasts with a mindset of potential and expression.”

-Andrew Timko

A recent example that showcased the benefits of experimentation was with the 2019 Worka — Legese Lemiso. It was an Ethiopian coffee from our partnership with MOPLACO. When we first started working with the coffee, we were keeping the yellowing phase, or Maillard phase, on the shorter side. This is more in line with how we typically treat brighter coffees like a washed Ethiopia. After a few roasts, we decided to try a different approach by extending the second phase, and we were amazed by the results.



Experimentation is crucial when we start working with a new coffee, especially one that comes from an importer rather than from a sourcing partnership. When importers send us samples, they come in small quantities. Small samples are enough to give a snapshot of a coffee before deciding to commit to it. Importers typically sample roast the coffee themselves before sending them to us. As a result, they receive an initial understanding of a coffee. Sometimes, they include flavor notes when they send it to us, which can help decide how we want to roast it. Occasionally, importers send us pre-roasted coffee if it possesses particularly unique qualities they want to convey. Mostly, they send green coffee.

When we sample roast, we only roast less than a pound of coffee. We use a smaller tabletop machine to roast samples instead of our Gothot Roaster. Sample roasting gives us an idea of what we can expect from a coffee. Working in small, quick batches allows us test out a spectrum of different methods to give us a fuller understanding of a coffee.

Once we sample roast a potential candidate, we collectively cup the coffee before making the final decision to add it to our roster. Typically, this decision is based on majority rule. During that initial cupping, we also do group calibration. This helps guide the flavors we ultimately want to develop in the coffee. We document our notes in Cropster. Once we decide to buy a coffee, it takes about 4-6 months to export from origin and import into the US. So, our documented notes on Cropster about the coffee and similar coffees help guide the roasting process. From that starting point, we try to mirror and replicate the flavors we achieved through sample roasting in a full production batch using our Gothot roaster.

“To me, there's actually a good amount of pressure in roasting. You not only want to present your company well, but you want to present the farmer's work in a way that you're proud of and you know that they would be proud of too.”

Mazi Razani

Mazi is careful and meticulous during coffee roasting, accounting for past experience and real time feedback from his senses.


When we work with coffee that comes directly from sourcing partners, we typically have a better immediate understanding of the coffee than we do with new samples. If we’ve worked with a specific coffee before, we have a good history of its flavor profile. This usually includes comments about that coffee from cuppings, roasting, wholesale accounts, and customers. More importantly, we will have a better overall understanding of the farmer’s practice, terroir, and intentions. Especially if we have done collaborative projects with that farm in the past.

When roasting a coffee we had in a previous year, we’ll look over any documented flavor notes and roasting methodologies we used. We use this information as a starting point for the new batch. Ultimately, we want to treat the new coffee as a unique product, but we utilize helpful data. Coffees can change pretty drastically from year to year. Sometimes a coffee tastes like a completely different product than it did a year ago. All of this is to say that no matter what, trial and error is integral to the roasting process.


Once a coffee we’ve purchased arrives at the shop, we do a quality assurance roast. We see how it reacts in the roaster. This lets us know what factors we need to change in order to get the full batch of coffee to match some of the elements we unearthed during sample roasting. From there, we’ll taste the coffee, make adjustments, and then it's usually ready for sale.

Sometimes that process isn’t quite so cut and dry. Because we are so excited by each coffee’s individual potential,  it’s hard to know when we’ve gotten the final roast exactly right. It’s easy for us to wonder if there are elusive flavor notes hiding. We ponder, "what if we could create an even more interesting composition?" We’re also balancing and taking into consideration the opinions of the entire team, wholesale partners, and café customers. Ultimately, a coffee is ready when it reflects the farmer’s hard work, and when we feel that our customers will enjoy it.

Mazi considers the partnerships he represents with each roast.

“Because we're in partnerships with so many people, I never want to feel like I'm roasting coffee just for me or how I want it to be, but rather how we collectively envision presenting coffee. I have to make sure I'm meeting those marks and producing something that is quality.”

-Mazi Razani


Once the coffee is on our shelves, we aren’t done trying to perfect it. The goal is for the coffee to get better as we sell it. Our meticulous quality control and assurance process allows us to continuously record feedback, ensure the best quality, and tweak the roasting method from batch to batch. We cup every single one of our roasts. Each batch either passes or fails depending on if it has met our standards. We also visually measure the coffee based on its coloring. Generally speaking, too dark of a brown may suggests overdone and bitter coffee. On the other side of the spectrum, too light a color may present a jalapeño or green bean flavors.

We typically let coffees rest for at least 24 hours after roasting before we cup them. Since we ship the day of the roast, if we later find that it doesn’t meet our standards, we contact customers and request they send the coffee back so that we can replace it. We document, store, and compile quality problems and all feedback from partners and customers in Cropster, which also has a database feature within the software. We can then synthesize and analyze all of the feedback and data to help improve future batches of a coffee.


Every roaster and roasting company wants to believe they are achieving the ideal expression of a coffee. Realistically, that is a very specific and personal perception. Throughout the roasting process, we juggle and balance not just the complex chemical and physical variables of coffee, but the opinions of our customers and our commitment to quality. We’re also taking into consideration the past and future of industry roasting practices. They help us decide what our continued point of view in the coffee world will be.

When we started Blueprint Coffee six years ago, we made the decision to roast out of our first café space. At the start, we roasted about 600 pounds of coffee per month. Along the way, we have gone through a range of lessons and growing pains. Ultimately, we learned how to be extremely efficient with our space. Now, we’re roasting 1,500-2000 pounds of coffee per week, and the beauty is that we can still pause in the middle of a roast to answer the questions of an interested customer. As scientific and technical as roasting is, it also represents the collective efforts and experiences of our cafe.

“All of these variables make roasting a complex, frequently frustrating, and rewarding all at the same time. Can anyone ever honestly say they ‘captured the essence of the coffee bean?’ I would argue no. At best what we can do is deliver a balanced expression of a coffee, which is up to us as a company. After that, it's about delivering a shared experience through coffee.”

Andrew Timko

The start of finding what a coffee has to offer is on the sample roaster.