Gesha seeds at a farm in Acatenango, Guatemala
Feb. 22, 2018
A common question we get from customers about Gesha coffee is if they're really worthy of all the fuss and high prices. That's a tricky question to answer, since the intersection of a coffee's taste and value are completely subjective. A nice wet-processed Gesha coffee can show characteristics of citrus and delicate tea notes, and not to mention express overt jasmine-like florals, qualities once thought to be found mostly in coffees from East Africa. That is, if you don't roast the heck out of it. So rather than tell customers whether or not they'll like Gesha, I think first we should look at how a properly developed roast can coax the "Gesha" qualities from the lot you've selected in order to have a fair shot of making this decision yourself.
I should preface this by saying, the roast ranges we recommend in our reviews of our Gesha coffees differer slightly. Most we don't recommend taking beyond City+ because much of what makes them "Gesha" in flavor is lost with roast development, while others retain floral and fruited tones on up to Full City. But one thing they all have in common is their ability to shine in light roasting. So for this blog, I'm testing light roast levels only, using our Guatemala Acatenango Gesha as our test coffee.
The Behmor 1600 Plus is as good as any of the home roasting machines we sell in that they're able to roast 1/2 lb of coffee in a relatively short time. By "short", I mean 1st crack in 10-11 minutes, and with enough charge to finish a City roast in roughly 2 minutes more. This is important to consider, as we've found that small-batch roasts dragged out much beyond this minute territory tend to have flatter acidity, lower sweetness, and all around less dimension.
I ran two test batches back to back, my goal being to reach the same amount of moisture loss for both, but with one roast taking at least a full two minutes longer to finish. This meant ramping up the heat at different rates, and especially for the slower roast, making sure to apply enough heat to keep the roast progression from stalling out. Below is my minute-by-minute roast log tracking internal chamber temperature, any adjustments in power/heat, and 1st crack (in green) and finish times.
roast charts for both roasts - temp readings are from the internal chamber temp which you can see by pushing the <B> button while roasting in manual mode
Given that my goal was to compare similar roast levels with drastically different roast times, I'm happy with how these two roasts turned out. The roast levels between A and B only differ by about .5%, but roast B has almost 2 full minutes more time in the roaster.
Both roasts batches were 200 grams of coffee going in, roasted in manual mode, and with high speed drum rotation. Roast A started out with heat full blast (P5), dropping heat to 25% (P2) when the coffee nears 325F, and letting ride out to the finish. I ramped the heat up more slowly for Roast B, starting off at 50% (P3), up to 75% (P4) at 3 minutes, and 100% (P5) at the 7 minute mark. This roast still neared the dreaded auto shutoff/overtemp error that occurs at 325F right around minute 12, and so it was necessary to drop down to P3 in order to avoid botching my roast. Thankfully it had just enough charge to power through to the finish without any more heat-setting tweaks.
Looking over the minute-by-minute progression you'll see a big difference in rate of rise (ROR) between the two. For the first three minutes Roast B is progressing at less than half the rate of Roast A, slowly catching up around minutes 6 and 7 after both heat settings are at full power. It's worth noting that after the exhaust fan turns on at 7:30 tracking ROR gets a little dicey, as the roast slows down a touch with the cycling of the fan. Roast B for example, you'll see the temperature trajectory go back and forth between minute markers 10, 9, and 8. I believe that has to do with the fan shutting off momentarily causing the roast chamber temp to jump by a few degrees. (my temperature readings are from the internal Behmor chamber probe, displayed on the LED by pushing button <B> during manual-mode roasting).
a simple but effective cooling tray made with a box, collander, and shop vac
I'll also mention that I cool my batches in an auxiliary cooling tray. I showed this setup in a previous Behmor roast blog, but basically I cut a hole large enough to set a small collander in the top, and then cut a small hole on the side in order to insert a shop vaccuum tube. I don't like how long it takes to cool coffee in the Behmor, and long cooling times can be linked to lower perceived sweetness in the cup (my next Behmor blog will be about this). I'm able to completely cool a 200 gram roast batch in my cardboard box setup in under 3 minutes.
OK, so now that you're familiar with the two roast profiles, let's have a look at what a .5% difference in roast development looks like, and more importantly, whether or not there's a difference you can taste. Judging the roasted coffee physically, you'd be hard pressed to notice any roast color variance in whole bean or the ground coffee. I expected Roast B to look slightly darker given the longer roast time, and potentially more internal roast development (though the weight loss was lower than Roast A), but that's not the case. I also did not see any difference in crust size when pouring the hot water, another visual marker that helps compare roast development to a degree. If one roast is darker than the other, it should have a thicker and more frothy crust than the lighter roast.
side by side comparisons of my two roast profiles
Cupping them side by side, there were noticable differences in levels of sweetness and acidity. I expected the longer Roast B to have flatter acidity and less floral notes, and it did. Knowing that I was biased going in, I had someone set the coffees up "blind", with nothing marking which was which, and I still picked it out pretty easily. Don't get me wrong, it was a thoroughly enjoyable coffee, but I found it to have a doughy flavor to it, more muted sweetness, and the floral quality was much less intense.
Roast A definitely had more overall dimension, convincing sweetness, and floral and light citrus notes were more easily discernible. It also showed a slight bittering roast tone in the finish that was not present in Roast B, but in my opinion, didn't take away from delicate top notes. And while I picked this one out as my favorite, and showing the most "Gesha" cup characteristics, a couple staff members were drawn to the other, part of this division resting on the aforementioned bittering roast tone.
Neither roast were perfect, but honestly, both were downright delicious brews. I personally liked Roast A better, and think I would try to improve on this roast profile by shortening the overall roast time even more by first tapering my batch size and adjusting from there.
To sum up, the easiest way to unlock the unique cup characteristics of Gesha coffee is to roast on the lighter side. That is to say, the floral, tea, and fruited top notes - what much of their value is tied to - are all but lost when roasted too dark, which for some of the lots we carry is anything beyond City+. This coffee's intensity is also dependent on overall roast time and jeapordized when roasts are stretched too far. If you're looking for a complex coffee that holds up to dark roasting, I recommend Kenya or maybe dry processed Ethiopian coffees. But for those who enjoy the vibrance that comes with light roasting, Gesha coffee's present a unique opportunity to try a coffee from Latin America with inherent floral and tea characteristics that are unlike any other.
***I want to add that while florals are muted in Full City brews, they do show up when brewing as espresso. Personally I think $18-20 per lb is a steep price for green coffee to use as espresso, but we do have customers who buy our Gesha for just that purpose. In an effort to come up with a more affordable way to enjoy Gesha as espresso, we've been fooling around with blending, and came up with a delicious blend of Gesha with a couple nice washed African coffees. Look for it in the next couple of weeks.