Let's have a look at how to view the test results and understand them. You can either watch the short video walkthrough below, or if you prefer to read you can find the step-by-step written guide below.
Two types of tests
The Verify Meridia portal conducts 2 types of tests to assess EUDR compliance:
Compliance Tests
Data Quality Tests
Compliance Tests:
These tests currently assess specific data aspects to indicate the risk of (non-)compliance to the EU’s regulation for deforestation-free products (EUDR). The tests cover different aspects, from deforestation, protected areas, land rights, and more.
Data Quality Tests:
These tests are designed to assess data quality. Results from quality tests can be used to gain insights into the general reliability and trustworthiness of a given dataset. Depending on the purpose of the dataset, different criteria for assessing data quality can be applied. For example, for EUDR compliance testing, particular focus is put on aspects of the data that pertain to farm location and farmer authentication.
Viewing and understanding Test Results
How to view test results
In order to view your test results for a dataset you have uploaded, go to the ‘Dataset’ page (the very first page that opens when you login), and click the black button ‘See Results’.
This takes you to a dashboard with a high-level overview of the results.
The question the system is trying to answer is: How risky is this dataset from a compliance and data quality perspective?
The output of each test is a risk severity, ranging from low risk, over medium and high risk, to critical risk. These will apply to specific farm plots.
What do low, medium, high or critical risk test results mean?
Low risk: From a data quality perspective, this means that tests have identified no material data quality issues and from a compliance perspective they have not detected anything that indicates non compliance issues with this farm record.
Medium risk: From a data quality perspective, this means that tests have identified issues with this farm record. These issues can hamper the accuracy of analysis but do not cause a material decrease in reliability. From a compliance standpoint, it could be weak evidence that issues with compliance could be expected. It is unlikely that these issues are substantial. Nevertheless, we advise investigating further.
High risk: From a data quality perspective, this means that tests have identified substantial issues with this farm record, resulting in a serious decrease in reliability. Accurate analysis is not possible. From a compliance perspective, tests have found strong evidence that this farm plot would not be considered compliant. Immediate investigation is advised.
Critical risk: From a data quality perspective, this means tests have identified significant material issues with this farm record. Farm records should be invalidated and excluded from the dataset immediately. From a compliance perspective, this means that tests have identified significant evidence that this record is not compliant. Immediate action should be taken to exclude produce from this farm plot, notify relevant parties, and start an investigation.
Invalid Result (NA): From a data quality perspective, if a farm record is missing data necessary for a specific test, it will receive an NA (invalid result) for this test. The missing data should be addressed. The urgency depends on the type of data missing (e.g., missing geometry is more urgent than missing mapping date). From a compliance test perspective, if a farm record receives an NA in compliance scores, it had serious data quality issues which made an analysis impossible.
If you prefer to view these definitions in a more visual way, below is a table view that shows the split between compliance tests and data quality tests per level of risk:
Example: Understanding the Results Dashboard
Let’s dive into an example to see how we could interpret the results from an uploaded dataset. You can either watch the short video here or read the written guide below.
At the top, blocks display a quick sense of the dataset's risk profile. For example, for this dataset 83.33% of farm plots have a critical risk. (That's very high. There are some high-risk, medium-risk, and low-risk plots. This would be cause for concern if you uploaded this dataset.)
Additional information includes:
Number of farm plots: 30.
Commodity: Coffee.
Origin: Indonesia.
If multiple datasets were uploaded during the upload phase, you can switch between them at the top of the dashboard.
You can download the high-level results for the full dataset as an Excel file or GeoJSON.
Filtering Results
When uploading data, you have the option to create filters, and this is where they come into play.
In this dataset, filters include aggregator and mapping date. We can toggle between these different filters to see different perspectives on our data.
For each aggregator, the percentage of critical risks is displayed.
Clicking on a column sorts aggregators from highest to lowest risk.
For example, if you’re interested in the aggregator Malangsari, which has 80% critical risk, you can:
Click the black button 'More Details' for this aggregator to get an idea of which tests are critical, for example.
Drill down to specific farm plots with issues. For example, LGL01 - Farm plot overlaps with protected areas.
If I click the white 'View Farm Plots' button, this will take me to see all farm plots for this dataset that are failing this test. I can scroll down to view all of them.
Use the Query Builder to get Granular Insights
You can explore your results even further by using the query builder. There are quite a lot of fields that you can choose from and you can apply multiple filters at the same time to really view something specific or to see what the overlap is and try to spot trends.
Here we can see that there are 5 farm plots failing for LGL01 - Farm plot overlaps with protected areas for the Aggregator, Malangsari.
If you scroll to the right in the farm plots section you can view compliance scores (e.g., critical) and data quality scores. You can also click the black button 'Test Details' to get further details.
This level of detail will show you this farm plots test results for all of the tests that were run by Meridia Verify.
Why are data quality tests important?
Data quality underpins EUDR compliance. Errors such as incorrect geolocations, topological mistakes, and implausible plot boundaries can derail submission efforts. Data quality issues can range from minor discrepancies to significant risks that could compromise reliability. Additionally, datasets with doughnut-shaped plots or other anomalies will face automatic rejection by the EUDR Information System (where due diligence statements under the EU Deforestation Regulation must be submitted).
You also have the option to export this data in an excel file or GeoJSON to explore these test results further and start applying resolutions.
Conclusion
In conclusion, understanding and interpreting test results is crucial for ensuring EUDR compliance and maintaining high data quality standards. By leveraging tools like the query builder and applying insights from compliance and quality tests, users can confidently address risks and prioritise key areas for action.