Skip to main content

Cafe Imports’ Controversial Roast Profiles: Criticism and Response

An important and certainly interesting dialogue has developed since green coffee importer Cafe Imports announced a new program that shares the company’s in-house roast profiles for specific coffees.

The program is an extension of what Cafe Imports terms Beanology, an informational guide to roasters on each of the coffees it sources. The company has partnered with roasting software provider Cropster, which provides the charts on each of the Beanology profiles after work with Cafe Imports’ own Probat L5 in Minnesota. Click here for an example of a the Beanology roast profiles.

cafe imports roast profiles

An example of a Cafe Imports roast profile using Cropster software

Praise for and criticism of the program have come in many forms, and I will not attempt to cover all of them here. But some of the heaviest criticism of the program came on August 13 from James Hoffmann of London’s venerable Square Mile Coffee Roasters. On his personal blog — which happens to contain a wealth of interesting content on all kinds of coffee industry issues — Hoffmann leveled some more technical criticisms:

Probe type, probe placement, probe depth in drum and percentage of full load will all yield different temperatures on a display for roast’s actual temperature.  We call them bean probes but that’s mostly a lie.  They very rarely are giving us an accurate picture of the bean temperatures – you can get pedantic and argue that all roast graphs should start with the bean temperature at room temp, instead of having a 60-80 second period of decrease before it bottoms out and starts to increase again.

Even the general shape of a profile – something you could argue is mappable from Cafe Imports’ roast profiles – will be different on different roasters.  I don’t really know how to translate the profile from a full batch 5kg Probatone to a Loring running a half batch, let alone to something like a Sivetz fluid bed roaster.

As well as some more philosophical objections:

The demand for roast profiles makes me a little sad.  It is a stark that people are buying coffees just to fill holes in offering lists.  They just need a Guat, or El Sal.  They want one that is traceable, with a good, saleable story (nothing wrong with that, I should add!). They want a shortcut to make sure they get reasonable value for money, but in many ways it is an admission that cup quality isn’t the most important thing.  You might even be more successful working this way.  Focusing on selling a story, a product, a service rather than just focusing on the experience in the cup.

To his credit, Cafe Imports’ Joe Marrocco took the time to write his own well-reasoned response to the criticisms put forth by Hoffman and others. Much of Marrocco’s justification for moving forward with what has quickly proven to be a contentious program relates to building a sense of collegiality and shared learning in a roasting industry that has historically felt somewhat divided, whether for proprietary reasons, or simply due the nature of smaller business. Here is a section of Marrocco’s response, in which he addresses roasters:

We do not want to tell you, our clients, how to roast. We do not want to tell our partners who are growing coffee how to manage their farms and process their coffee. However, we do want to be your partners and help in any way we can to set you up for success. We do not wish to insult you, or unveil any trade secrets. We want to encourage communication and collaboration to spur on innovation in coffee roasting. We want you to see the intent with which we approach turning these green beans brown internally and the power of sharing experiences and building community.  This is meant to be collaborative and start discussion.

I don’t want to strip too much from the original posts written by Hoffmann and Marrocco, respectively, if only because both are worth a read in full.

Click here for Hoffmann’s full post

Click here for Marrocco’s full post




Good piece. Thanks for posting it.

When Cafe Imports first announced this, I kind of smelled “gimmick” a (square) mile away. But the history of business is filled with gimmicks — some of them successful.

As a home roaster for over a decade, I’ve wallowed through enough of my own share of temperature profiling instrumentation and data analysis. In the end, the best that has ever come out of that has really depended on the variances associated with the bean — akin to Tom Owen’s old “Stretching Out The Roast” series where holding certain temperature profiles can influence the end results.

Of course, you can rightfully argue that a little data is always better than no data. And for that, at least Cafe Imports is trying. But is it useful data? Far from it at this stage I’d say. It’s still no replacement for the experimentation required with a new roast to dial in the characteristics you like best yourself.

Rich Westerfield

Doesn’t bother me in the least. As Joe M. noted, he’s not telling anyone how to roast their coffee, just telling you how he roasted the coffee. There are still hundreds (if not thousands) of roasters who don’t profile at all. I apprenticed at one of those and eventually built profiles for them (which they ignored after I left). I like looking at profiles from others, gives me a better idea what they’re trying to accomplish regardless how I choose to interpret that particular bean and I learn a bit about roasting equipment effect on a bean. I am a “can’t have too much information” type since there’s such a dearth of easily affordable & accessible good info roasting chemistry out there.

Chad Kimm

I’m trying hard to figure out how this can be considered controversial. To me this is really just an extension of the cupping notes. This profile roast, combined with tasting results from a sample batch roasted in a common sample roaster, give a more complete picture of how Cafe Imports came up with the flavor descriptors that they publish in their Beanologies.

I certainly don’t look at this as a recipe for how a particular coffee should be roasted. That is completely up to the person behind the roaster.


I have a nose, I have a mouth, that is all I need. Such vanity graphology is simply silly–and I say that as someone who does statistical analysis for a living.


Bryan (Aug 30, 20:38)

Pull your head in. Cafe Imports are not telling you how to roast, but merely sharing the profile they used. You are free to roast as you like. Quit the self importance and just enjoy the coffee.


Whether you profile or not, it is a good idea to keep your mind open to how others approach the same bean. To that end, I like the program. If you want to follow the profile exactly, that’s your choice, too. Calling it a silly gimmick will not make it any less useful or educational.

die Rösterin

James comment is not a critic, it’s just a fact, but it’s sad that it has to be mentioned cause some wanna-be-roasters out there really think it helps getting a profile from someone else….. a good roaster DOES know that he cannot compare profiles with another one….specially not when they have different roasters. Even if we have the same machine, same size, same coffee – humidity&temperature makes it a hassle to repeat. Anyway, I like sharing knowledge with other roasters, too, like talking about bean behavior, and even curves – but basically analyzing what adjustments, temps and timings effect which flavors in the bean. So you should talk about chemistry, physics and what you want your result to be – but not just copy gas adjustments of other roasters. For example – I can play with my drum speed and the exhaust air. But my roaster sucks the air out of the drum with an external motor. The Genio roaster blows out the air of the drum – so that effects the air pressure in the drum at different temps. Even drum speed affects acidity and body – but is the drum speed of my roaster the same as on the genio? No, it’s for sure not. And our drums have a different radius, too. But what I love about my cropster is, that you can visible compare several profiles at the same time and see immediately why coffee #1 tastes different then #2 and #3….and you can use the good profile as referenzcurve the next time roasting. But you will see, that in order to follow the same profile curve you need different adjustments EVERY time 😉 So profiles are good for your own work but don’t help others. Still it’s ok to compare with others, but just in order to learn how much different another roaster has to roast because of his machine 😀 I like that Cafe Imports shares their knowledge, but one has to be able to read it properly and don’t get intrigued by just copying – that wouldn’t work 😉

David Palmer

I don’t get it, really. I’ve come into the business with a thirst for knowledge, standing in a hallway of closed doors. Honestly, the Roasting business is over protective, condescending and blind in so many ways. How something as simple as a Beanology could infuriate so many people stands as an idle demonstration to why Folders, Maxwell house, Starbucks and Green Mountain have been able to steal over 95% of the greater Coffee market from both consumers as well as the Farmers. To be clear, I’m not a genius, but I think it goes without saying my Fluid Bed Roaster isn’t going to profile like a Probat. That said, I think it’s quite awesome to see a multitude of Beans as they perform and all the variables within the process laid out. We all know, there are “secrets” to every Roasters process that make their Coffee superior to everyone else’s, but this? This is the sharing of science, and from what I’ve come to know of Cafe Imports, they’re big fans of that. Salute.

Comments are closed.