Let’s talk about traceability….help us solve this problem

In case our coffee wasn’t enough to get your brain moving in the morning, here’s a traceability teaser to wake up your brain. We are also here to enlist YOU to help us solve this on-mountain problem related to traceability of coffee when it is sold in parchment form. After reading, please submit your ideas in the Comments section of this blog post.

This scenario is based on a real situation observed just last week in the greater Mbale region of Uganda.
First, allow me to introduce the two sides of the traceability argument.

You are a farmer. You pick your coffee berries and have two options. One, you can carry your cherries to the nearest washing station, where you will be paid per kilogram on the spot. Alternatively, you can pulp the coffee, wash and dry it to “parchment” stage at home, store the parchment in your bedroom until the price is high or until you need the money, at which point you can call a local buyer to come by and pick up your parchment. While pulping, washing and drying requires more work, at least this way you don’t have to carry the berries all the way to the washing station (these berries weigh 5 times as much as the dried parchment, which can be picked up at your house). Plus, holding onto the parchment acts as a bank; you can cash-in whenever you feel like it.

You are a coffee exporter. You can buy coffee in the cherry, allowing you to grade quality at the point of purchase (red = good, green = bad), wash and dry consistently for consistent quality. However, the buying season is limited; if you don’t buy parchment as well then you will fail to meet the tonnage you require to fill your customers’ containers.


In an effort to increase the quality of coffee purchased, an exporter invests in a community washing station. Drawing from a model which has seen success around the world, the importer turns operation of the washing station over to the farmer community, who can then control their crop one step further up the supply-chain, from cherry to dried quality parchment.

However, out of habit, each individual farmer begins using the washing station to pulp his coffee into parchment, and then takes the parchment home to store until he needs the money or until the price for coffee rises. The farmers say that they’ll bring the parchment back to the washing station when they are ready to sell, but in the meantime they are using the pulper and washer for free.

The exporter, on the other hand, is concerned that the parchment which is brought back to them at a later date is different than the coffee which is pulped and washed in a high quality manner at the washing station. For the exporter to pay premiums for premium coffees, they need to know that the parchment they purchase has been processed well. However, from looking at the parchment there is usually no way to know that the farmers are not selling the parchment processed here (high quality due to washing station precision) to different exporters or middle men on the mountain at a premium, only to bring lower quality parchment back to the washing station to sell at equal (or higher) prices AND receive the “second payment” bonus that the washing station owner has in its contract with the farmers.

The natural conclusion for the importer was to take back control of the washing station, buy only red cherries and not allow farmers to bring in any parchment.

Is there any way to alleviate the concerns of the importer while keeping the washing station under control of the farmers, who often want to sell parchment?

11 responses to “Let’s talk about traceability….help us solve this problem

  1. You have traceability to the grower level, with each grower being assigned a unique number. You have that growers parchment rated for quality in a publicly accessible database. A grower with consistent high quality will continue to the the orders. An inconsistent grower, or or one selling low quality, will either not sell at all or sell at reduced prices.

    The growers could have a “reputation score” that went along with their parchment so that the buyer would be reassured of that grower’s reputation in the community.

    How you do the electronic database is the question. It can be done cheaply, but I won’t go into how because there are many ways to go.

    • John –

      Thank you for this……very, very relevant.

      Farmers ARE currently registered, a database (mostly handwritten) which includes counting the number of trees they have, thus providing the exporter an idea of how much coffee they should be able to bring, preventing farmers from getting lower grade coffee from the lowlands and selling for higher prices at high altitudes.

      However, the novel concept you introduce here is that which is borrowed from online commerce sites such as eBay – the reputation score. I do believe that there is a lot of merit to this concept, and that if financial incentives were given out to farmer groups for the collective quality of their coffee, then this would create a dynamic whereby farmers WOULD enforce each other.

      However, there are two issues that remain….one logistical and one psychological. Perhaps with a bit more thought we can solve these too.

      Firstly, quality can be assessed at two points. (1) when the coffee is brought in as cherry. In the scenario described above, this cherry is converted into parchment, which then the farmer takes back home with them from the washing station. So while there is traceability at the farmer level, the traceability of their coffee in parchment form is no longer relevant because high quality cherries can come into the station, and high quality parchment would be taken away, but no one is sure what the quality of coffee brought back to the station would be later in the season. You see, farmers would be able to sell the washing-station processed coffee to anyone, and then take different parchment back for selling to the washing station later on, undermining even farmer-level traceability. The second point at which coffee quality can be assessed (2) is after parchment is dried, hulled, and then cupped. The problem in assessing individual farmer performance here is that the processing is necessarily done in mid-sized batches…..whereby a farmer may sell 10 – 15 kilos of coffee at a time, the processing is done at larger volumes, requiring many farmers from a given farmer group to contribute.

      And this leads us to the second problem, which is that we are dealing with a little bit of game theory here. Once in parchment form, the only effective way to evaluate coffee is by producer organization or farmer group, not by individual farmer (remember these are small farmers, 200 – 1000 trees per farm).

      I think that, over time, a reputation score can help to provide a mechanism for the internal regulation of farmer groups vis-a-vis this game theory problem, and that this concept can be baked into many purchasing / transparency models – so thank you for this relevant contribution.

      – jake

  2. Pingback: Uganda sourcing trip – photo dump « Crop To Cup's Weblog·

  3. What if the washing station also had a storage facility where each farmer who used the station had a storage room with lock and key in which to keep their parchment until they were ready to sell? It could be as simple as concrete and doors with padlocks. As long as security were maintained, the farmer wouldn’t have to haul the coffee back and forth from the exporter’s site. Furthermore, the exporter could use a per bag rating system (like a barcode or number on each bag that’s stored) for future reference when the bags come out of storage. Probably security is the biggest issue here…

    • Good thinking Amanda. With a large enough space all of the coffee could be dried and stored properly in the farmers community. Taylor’s posts about washing stations in Burundi are big enough to dry and store the coffee, for example.

      However, you have also uncovered another underlying issue faced by Washing Stations in Uganda – scarcity of land. The mountains are scarce for open land in the first place, and as individual investments washing stations are built on small plots. In that a single station should look to have collected 40+ tonnes of cherry (32+ tonnes parchment) this far into the season….which is simply too much for them to store and dry properly without taking them down-mountain where there is more land, more sun, and more drying beds. Maybe if more farmers pitched in by putting up their land into some sort of commons the washing station could be bigger. But then again this deals with redistribution of land….which can be a touchy subject sometimes.

  4. Crop to Cup.

    I have a solution that would work:

    1) Place an RFID chip in the package
    2) Have someone at the washing station sign (using a sharpie) over the seam in the wrapped parchment package so that if it were opened, it would be apparent.
    3) Have the scannable RFID connected to a database with all of the information about that parchment.
    4) Collect the RFID chips (they probably cost 50 cents each and return them to the washing station when the parchment was sold.

    You would need GTINs to do this but our company would be willing to donate some to the cause if you contact us.

  5. Jake,

    If the RFID were too expensive, you could use databar labels. These cost less than a penny each and you could sign with the sharpie and put the databar label over that signature, and then cover the databar with clear plastic tape.

    The Databar would have a human readable number on it but the GTIN encoded in the databar would not be readable except by a scanner. This way the farmers could not make forgery databars and the ratings of the parchment could be done with an iphone or a google droid that was capable of scanning the databar and taking ratings using shopsavvy’s free software.

    This would mean one smartphone per washing station and 8/10 of a penny per parchment would be your total cost.

  6. I think the technology approach is a good one. In the end, I bet that solutions like these will reinvent the coffee processing scene. Progress is slow, slow slow, however – we still have challenges using even SMS technology with farmers on the mountain. Still, its good to see progress, and in a half year or so we’ll probably launch our first big farmer technology program.

    I think the RFID and Databar approach will be key here – John, let’s definitely be in touch about this. For this specific issue, however, the coffee cannot be sealed as it leaves the washing station. After farmers pulp their coffee at the station they return with it to their homes to start the drying process, i.e. it can only sit sealed in a bag after a week or two or more of drying under the sun.

    But I think we are making progress here! Ratings will also be key, especially since we have seen clearly on the mountain that many quality-focused farmers do not contribute to the whole since they are worried about the lower quality of others bringing down the combined crop.

Leave a Reply to John Bailey Cancel reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s