More and more global brands and their suppliers have been impacted by the requirements set forth in the Dodd Frank Section 1502, regarding Conflict Minerals. The regulation stipulates that publically traded companies must disclose their Conflict Minerals sourcing. In order to meet this requirement, it is critical to: a ) use best practice learnings from existing SEC conflict mineral filings, b ) properly analyze the data, and c ) ensure dependable data interpretation to optimize your Conflict Minerals program.
To achieve best in class analytics, Source Intelligence draws on the expertise of our team of data-driven compliance experts. To communicate some of the aforementioned best practices, we turn to Data Scientist, Eric Lessert, to provide insights on interpreting Conflict Minerals filing data analyses. “Data analysis should be a holistic approach. There are many factors to consider. Overlooking details in methodology can lead to incorrect interpretations,” commented Lessert.
According to Lessert, any summary report should be based on the following:
- Choose an appropriate visual representation of your data. This will alleviate any confusion in your conclusions which may be caused by the presentation format. For example, a three-dimensional bar graph may lead the viewer to overestimate the provided values, whereas a two-dimensional bar graph provides a clearer view of the data.
- Provide clear definitions for the values or labels used in graphs or tables to define their significance. For example, “high”, “medium”, and “low” need to be accompanied by raw data that clearly defines these labels.
- Ensure the appropriate range and representative value for the variables are understood. If the averages are stated without the range, it becomes difficult to determine any sort of context.
- Select the appropriate analysis or model based on various technical prerequisites. To communicate the stated conclusion and to make the analyses are more transparent, provide the supporting rationale for the use of the chosen analytical model.
- Cite the data source and any filtering or changes intended to clean it up, if any. Doing so allows the viewer to verify the conclusions through replication of the analyses involved.
- Provide the raw data where possible so that viewers may perform their own analyses if they wish.
- Finally, if a statistical model is used, the level of error in the measurements should be included. Without an error range, interpolation or extrapolation becomes murky at best.
Interpreting existing SEC conflict mineral data correctly is critical to accurate benchmarking and the meaningful implementation of best practices into your future Conflict Minerals program. Involving a data scientist in the process is paramount to a successful reporting year.
To find out how Source Intelligence’s analytical platform and regulatory compliance experts can better suit your compliance needs, click here.