Lessons from a 2 Day Coffee Crop Tour in Brazil

With a country as large and as important in the coffee world as Brazil, it should go without saying that you cannot get an accurate picture of the coffee crop in two days. Nonetheless, it is essential for members of the industry to continue their education with regular visits to origin.

The most important reason to visit origin is a human one, to maintain the important relationships. In addition to the human element, origin trips are also essential in keeping us connected to coffee as a biological plant, rather than simply a financial commodity.

In this article, I will share the biggest takeaway I had from my brief crop tour, what I think the keys are to a successful crop tour, and the differences with crop surveys.

My 2 Biggest Take-Aways

I won’t keep you in suspense, the two biggest take-aways from this trip was

1) the enormous potential of the 23/24 crop and

2)the variability in the 22/23 crop.

We will come back to these two points, but before I share the details of the tour, lets provide some context on what makes a successful crop tour.

What Makes a Successful Crop Tour

A successful crop tour requires understanding the limitations.

When doing a crop tour for a couple of days, or even a week, you can be little more than a tourist. Sure, I spent several hours over 2 days out in coffee fields looking at trees and examining the crop, but the amount of area that I could cover in that amount of time was insignificant.

It would be dishonest for me to make big pronouncements on the crop based on solely on these observations.

Here's what we can use direct observations for on a short trip:

  • Examples: Extremes, Mean, and Variance

  • Questions

  • Supporting Evidence

Finding Examples: Extremes, Mean, and Variance

For those of us in the coffee industry, one of the key purposes of doing crop tours is to build mental references. It is one thing to read about devastation to a crop from frost and drought, to see statistics of a bumper crop, and it is another thing to see these things in person. To hold the cherries in your hand, to look at the nodes on the branch, to feel where the beans are growing and not growing on the branch, to get stung by a fuzzy caterpillar.

Actually, I wouldn’t recommend getting stung by a caterpillar. You can scratch that last one off your list.

However, these mental references are important not only because they deepen your understanding of the biology of the plant, but because they provide a common language for discussing the coffee plant with other coffee people.

As mentioned earlier, coffee people are some of the most valuable resources that we have in the industry. Trips to origin help to develop and deepen these relationships, and they also allow us to "calibrate" over the coffee trees.

Many roasters visit exporters (and importers) to calibrate while cupping coffee. This enables them to both be speaking the same language when talking about sensory analysis.

For those of us who are discussing crop size and the physical coffee market, it is equally important for us to "calibrate" on yield, planted areas and crop potential out on the farm. This also enables us to have a shared language of mental references when discussing what is happening on a regular basis.

When building mental references, we want to develop a diverse library. This means observing a variety of harvest periods: blooming, cherry growth, harvest. It also means observing a variety of locations and environments, and also observing a variety of conditions in various stages and environments.


Your diverse mental library becomes an essential tool on future crop tours because it allows you to compare similarities and differences with what you have seen before. This ability to compare and contrast with other crops enables you to ask the right questions when you are out on tour.

We might ask these directly of an agronomist, trader friend or farmer: Why is this farm different? What happened here? Why are these leaves discolored? Why are these beans brown? How well do these farmers care for their crops? Is x or y typical?

Or these might be questions that you bring back with you and require further research.

Supporting Evidence

Sometimes you may be going out to the field with a particular question or problem in mind, and one of the best ways to make sure that your tour is effective is to look for evidentiary examples that answer your question.

Evidence gathered in small amounts on a short tour is not going to be the definitive word on anything, but it does enable you to at least confirm positive existence of phenomena.

For example, one of the key features I wanted to see was example of damage from frost. I found several examples that fit this. My guide Virgilio took me to a location where he took pictures of frost damage last year, and I compared these pictures with photos that I took yesterday. He also showed me examples of fields where the trees were dug up and replaced with wheat and corn, and also examples of minor, moderate and severe crop damage. I was also able to compare this to areas that were not damaged by frost at all.

Difference with Crop Surveying

On this trip, my purpose was not to quantify anything, but rather to confirm examples and get a feel for the extremes and the mean.

There are a number of different ways to get a useful crop number for an origin, but my preferred technique is what I would call the methodical observation approach.

The methodical observation approach to crop surveying requires finding several key pieces of data, but most crucially: the total planted area and the yield per hectare (outturn would also be needed, but that is beyond the scope of this article).

Total planted area can often be received from agricultural agencies such as CONAB in Brazil, although you may need to adjust it depending on how reliable the figure is. Once you have area figures that you are satisfied with, you need to yield per hectare. These two combined (multiplied)will enable you to come up with a crop estimate.

However, getting yield per hectare is not easy, and this is where crop touring is often used. The method that I was taught is as follows:

  • Visit many farms in each geographical cluster of farms

  • For each farm pick a random location within the farm (it helps that the farms are essentially grids, two random numbers between 1 and 20 will give you an x and y coordinate to start)

  • At the random location, estimate yields of 10 proximal trees (front of one row and back of another row)

  • Record the yields for that particular farm

  • Average all of the yields for the region

In theory, the more farms that are evaluated in a particular area would provide a more accurate representation of the total farms of that region. Although there could still be heavy variation in the results.

Another method of obtaining yields is what Steve Wateridge of Tropical Research refers to as a “micro-survey”. To conduct this, you need to have relationships with multiple farmers in the major regions of the origin and ask them fill out surveys self-reporting on their crops at regular intervals. This can be a rapid, cheap and useful data point, but admittedly is much less accurate than a well staffed crop tour.

Conclusions from my Brief Crop Tour

Now that I have clarified what can be accomplished during a very brief crop tour such as I took over the last few days, let’s discuss the results.

My tour covered Minas Gerais which is one of the largest Arabica producing areas in Brazil and I focused on the variety of outcomes from frost/drought damage.

As mentioned in the first paragraph, there was substantial difference in outcomes from 100% damage to no visible damage at all. However, even intentionally viewing damaged areas it was easy to see that there was still a huge amount of coffee (as would be expected from Brazil).

It was also clear how there could be some substantial differences in views among the major tradehouses. Even if they were using the same crop estimation techniques, there would be a substantial margin of error just based on which samples were selected.

The other big take-away was of course the new growth for the 23 crop. I didn’t set out to explicitly observe this, but it was impossible not to from the trees on the farms. Every farm that we visited, we saw fantastic branch growth. The growth was quite good at the farms with little damage, but in many cases it was even better on the farms that had damage.

This is typical of the coffee tree. When the tree is unable to put energy and resources into cherry growth, it puts more into the growth for the following crop, and this was a constant theme throughout the farms that we visited.

Moreover we can cross this observation with other data, such as rainfall and reports from other attendees at the Guaruja International Coffee Seminar. Roasters, producers, agronomists and traders all agreed that the potential for the 23 crop was very, very strong.