The Blog

view all

433. The CRS Colombian Varietal Cuppings: Intelligentsia

Today, a conversation with Intelligentsia QC manager Chris Kornman and coffee buyer Geoff Watts regarding Colombian coffee varieties—the third weekly installment in my series on the CRS Colombian Varietal Cuppings.

Earlier this month, Chris and QC Lab Assistant Amanda Seaver staged the CRS Colombian Varietal Cuppings at the Intelligentsia Roasting Works in Chicago.  Six different Intelligentsia cuppers participated.  They did not have a decisive preference for either variety.

Intelligentsia’s most experienced and best-known cupper, the great Geoff Watts, did not participate in the CRS Colombian Varietal Cuppings but did cup all 22 sample pairs during the first Colombia Sensory Trial panel, also hosted by Intelligentsia in Chicago.  He preferred Castillo, and not by a small margin.


At the outset of the process, I asked Chris and Geoff about their perceptions were of the two varieties and their expectations for how each would perform.

Chris: “At Intelligentsia, Caturras from Colombia are perceived as being sweet, juicy, red-fruit forward in flavor profile, soft and clean. Castillos are seen as tart, full-bodied, vegetal and bitter.  My expectation is that there will be a small preference for Caturras.”

Geoff:  “I think about Castillo as a cultivar that will produce a decidedly inferior quality cup as compared with Caturra.  I have had some very poor experiences with Castillo during cuppings over the last several years, and have come to associate the variety with a degree of roughness and a less enjoyable, less articulate cup.  If pressed, I would lay some small odds on Caturra outperforming Castillo by a reasonable margin on average, with a few exceptions.”


CRS Colombian Varietal Cupping.

Chris, Amanda and their crew preferred Castillo in 11 of the sample pairs and Caturra in 11.  They awarded Caturra a slightly higher average score (84) than Castillo (83.7), but a slightly higher median score for Castillo (83.6) than Caturra (83.5).  Exceptional Caturra samples (88 points and above) narrowly outnumbered exceptional Castillo samples (five to four), while most of the sub-specialty samples (under 80 points) were Castillo (five out of six).

The Intelligentsia cuppers were remarkably calibrated.  The rate of agreement on preference was over 90 percent.  In seven of the 11 sample pairs in which they awarded a higher average score to Castillo than Caturra, the preference for Castillo was unanimous.  The same was true for nine of the 11 sample pairs in which they awarded higher average scores to Caturra than Castillo.

Chris was not surprised about the degree of calibration.  He was surprised there wasn’t a clearer preference for one variety over another among cuppers who participated in the CRS Colombian Varietal Cuppings.

Colombia Sensory Trial.

Geoff, on the other hand, had a very clear preference for Castillo.

  • He preferred Castillo to Caturra in 15 of the 21 sample pairs he tasted.
  • The mean score for Castillo samples (84.6) was 1.2 points higher than the maean score for Caturra samples (83.4).
  • The margin of preference was even greater for the median scores: Castillo 85, Caturra 82.8.
  • Castillo samples earned more scores of 88 and above than Caturra (8 to 7).
  • Castillo samples had fewer of the scores under 80 (5) than Caturra samples (9).

Geoff was “surprised and delighted” by the results.”I wasn’t surprised that Castillo was capable of producing a very good cup.  We’ve seen that happen before.  But I was pleasantly surprised with the frequency with which it did so, and the number of times it outperformed the Caturra from the same farm.”



These cupping exercises have helped Intelligentsia reconsider its position on Castillo, but are not likely to have an immediate impact on the company’s buying practices.

Perceptions of Castillo’s cup quality.

Chris: “In light of this research, any bias we might have previously had against Castillo will likely be regarded as more-or-less unsubstantiated.”

Geoff: “I can say that I’m definitely much less resistant to Castillo than I had been before the trial, and I feel a sense of relief about that.  I’ve buying a decent amount anyway, since many of the farmers we work with have been planting it for years now, but it has still given me some anxiety. Now I’m a little more at ease about its ability to produce quality.”

Buying practices.

Geoff: “We’ve got to be careful about extrapolating too much from a single trial.  There are just so many factors that influence coffee quality, so we’ve got to be careful about making assumptions or making logical leaps based on the outcome from a trial that is still missing some degree of control over important variables.  I see this trial as the beginning of a longer investigation that should eventually give us enough insight and confidence to make better decisions about recommendations to farmers regarding varietal choices.  There is still so much more to learn.”


This conversation with Intelligentsia is the third in a series of weekly interviews on the CRS Colombian Varietal Cuppings

<< Last week: Tim Wendelboe.


The Colombia Sensory Trial and the CRS Colombian Varietal Cuppings are supported by a grant from the Howard G. Buffett Foundation.









Leave a Reply

Your email address will not be published. Required fields are marked *

Note: You can use basic XHTML in your comments. Your email address will never be published.

Subscribe to this comment feed via RSS