1. Key Considerations

What is the purpose of assigning unique Farmer IDs?

A Farmer ID or Producer ID is a code typically generated by the operator (e.g., trader) or aggregator (e.g., cooperative). It is not the same as a national ID number (the citizen’s service number) or National Traceability ID (which can be assigned by a national traceability platform).

The Farmer ID enables consistent and reliable linkage of the physical volumes to individual farmers and individual farm plots. The Farmer ID also enables the connection between the Farmer Name, the Farm Plot, and the Aggregator, aiding in the identification of misallocation of farms and other traceability data issues.

What is a farmer aggregator and how should it be structured?

A farmer aggregator is a general term for one of the first points where produce from different farmers is collected and registered. What a farmer aggregator is differs depending on the commodity and country (regions).

Examples of a farmer aggregator include: cooperative, licensed buying company, district purchase station, and washing station. More examples can be found at the end of this document.

There are a couple of reasons why we advise organising your dataset using aggregators:

We use the farmer aggregator data to run specific tests that provide additional insights.
We also use it to disaggregate scores per aggregator to provide insights into aggregators.

The farm-level dataset should include a reference to the aggregator name/ID and aggregator location for each farm plot record.

What is the value of including Farmer Names?

Farmer names are not required by EUDR, so it is not mandatory to include them. However, we do recommend including them, as they can be helpful in identifying data quality issues.

Including farmer names assists in ensuring that real data is obtained—i.e., an actual farmer is linked to an actual plot. The Farmer Name field is most valuable in our cross-dataset tests, where we compare clients’ data against the farm plots in our data lake and/or against plots in national cadastral databases.

By including the Farmer's name, we can better identify whether such cross-dataset overlaps are conflicts or can be explained by other factors.

2. Requirements for Farm-Level Datasets

The data file should be structured according to the table below, with the attribute fields included in the geospatial data file as attribute columns.

File Organisation

Provide the data as separate files for each origin country and each commodity.
Structure the files in a consistent manner and apply clear filenames.
The geospatial and attribute data should be merged into a single file.
If needed, the dataset can consist of multiple files (e.g., for different districts, geometry types, aggregators), as long as the geospatial (polygons, geopoints) and attributes (plot ID, farmer ID, farmer name, etc.) are together in one file.

Tip: Avoid sending individual files for each farmer or farm plot.

Geospatial Data Requirements

A geospatial projection (also called Coordinate Reference System, CRS) does not need to be applied. The default WGS84 is sufficient.
Possible file formats: GEOJSON (preferred), Shapefile (SHP), KML, or XLSX (not recommended for polygons).

Tip: We accept datasets with hybrid geometry types, including a mix of geopoints and polygons. However, sharing both geopoints and polygons for the same farm plot should be avoided.

Attribute Data Requirements

Ensure the attribute data of all farm plot records (geopoints and/or polygons) contains the mandatory fields requested in the data specification table.
Apply clear attribute labels from the Requirements for Dataset Contents table.

Tip: Avoid merging different field information into one column. Separate and insert the attribute information for each field in the correct type and under the correct header.

Data Quality and Submission

Before sending the data, ensure that the geospatial files contain no errors or corruption. This can be checked by simply opening the data in the program of choice, such as GIS, to confirm it works correctly.
Share the latest data version containing all farm records for applying the tests.

Tip: Avoid sharing multiple versions that may create duplication of information.

Requirements for Dataset Content

2 Farm plot size estimate, refers to the area where the crop is cultivated. It does not necessarily include the total land plot owned by the farmer or producer.

3. Aggregator specification

For cocoa, typical aggregators are:

cooperative (coop), often in Côte d’Ivoire, Cameroon, Ecuador
traitant, often in Côte d’Ivoire
licensed buying company district, often in Ghana
licensed buying agent, often in Nigeria
town collector, often in Indonesia

For coffee, typical aggregators are:

middleman or intermediary
purchasing station
pulping/de-pulping station
fermentation, washing, drying, cleaning, and/or sorting station

An example of a supply chain of cocoa in Côte d’Ivoire. In this case, the aggregator is not defined as the first point of aggregation and registration (cooperative sections), rather, it is defined at the second step (cooperative warehouse).