Calculation Logic

See how Tracenable transforms incomplete EU Taxonomy disclosures into complete, reconciled metrics through structured accounting rules.

The EU Taxonomy dataset relies on consistent rules to transform raw company disclosures into complete, standardized metrics. Not every company reports every required data point, and disclosure practices vary widely. To address these gaps, Tracenable applies a clear accounting logic that respects the hierarchy of EU Taxonomy dimensions. This ensures that users always have a full, reliable set of metrics for each KPI and reporting year.

---

Hierarchical Structure of Dimensions

The EU Taxonomy dimensions are organized around two key hierarchies:

• Level hierarchy

  • Activity-level – disclosure broken down by specific economic activities.

  • Total-level – aggregated across all activities of the company.

Totals can be derived from the sum of activities when missing.

• Eligibility Category (or Screening criteria) hierarchy

  • A+B (Eligible and Non-Eligible)

    • A (Eligible)

      • A1 (Aligned)

      • A2 (Eligible but not Aligned)

    • B (Non-Eligible)

Higher-level categories (A, A+B) can be derived from the sum of their subcategories (A1, A2, B).

These hierarchies define the structure of the dataset. The accounting rules build on this structure to fill gaps, derive missing values, and ensure consistent reporting across companies.

---

What We Deliver

For each KPI (Turnover, CAPEX, OPEX), Tracenable delivers up to eight metrics per company:

• A activity-level

• A1 activity-level

• A2 activity-level

• A total

• A1 total

• A2 total

• B total

• A+B total

Together, these metrics provide a complete view of a company’s reported and computed taxonomy performance.

---

Accounting Rules

Tracenable applies structured accounting rules to ensure that every KPI is complete, consistent, and reconcilable. These rules respect both the Level hierarchy (Activities → Total) and the Eligibility Category hierarchy (A → A1 + A2; A+B = A + B).

Rule 1: Compute Missing Parents from Children (Bottom Up)

Higher-level (parent) values are derived directly from their children. For example:

• A activity = A1 activity + A2 activity

• Total (A+B) = Total A + Total B

• Total A = Total A1 + Total A2

• Total A = Sum of activity A

• Total A1 = Sum of activity A1

• Total A2 = Sum of activity A2

This guarantees that roll-up values (e.g., totals) are always available, even when companies only disclose granular activity-level data. It ensures consistency across levels and prevents gaps in reporting.

Rule 2: Infer Missing Values from Related Metrics (Top Down)

Missing values are inferred using known relationships between categories and totals— but only when at least one side of the equation is a company-reported value (not already computed by accounting rules). This avoids circular calculations and ensures that inferred values remain anchored in real disclosures. For example:

• Total B = (A+B) – Total A (if Total A is reported, not computed)

• Total A = (A+B) – Total B (if Total B is reported, not computed)

• Total A1 = Total A – Total A2 (if both Total A and Total A2 are reported)

• Total A2 = Total A – Total A1 (if both Total A and Total A1 are reported)

By enforcing this condition, Tracenable avoids compounding errors and ensures that every inferred value is tethered to at least one direct company disclosure. This makes the dataset more reliable, transparent, and audit-grade, since users can always trace key metrics back to reported figures.

Rule 3: Compute or Infer Absolute Values

When only percentages (relative values) are disclosed, absolute values are computed from totals. For example:

• Absolute A1 activity = relative A1 activity × absolute (A+B)

• Absolute Total A1 = relative Total A1 × absolute (A+B)

• Absolute (A+B) = absolute A + absolute B

• Absolute (A+B) = constituent absolute ÷ constituent relative

This rule translates percentage-based disclosures into absolute figures, allowing users to compare across companies and sectors. It increases the analytical utility of the dataset by ensuring both relative and absolute values are available.

Rule 4: Apply Boundary Defaults

Logical defaults are applied when disclosures are explicitly zero or not specified. For example:

• Total A1 = 0% if Total A = 0% with absolute value = 0 or not specified

• Total B = 100% if Total A = 0% with absolute value = 0 or not specified

These defaults prevent missing or zero disclosures from creating inconsistencies. They provide a reliable baseline when companies report minimal information.

Rule 5: Establish Baseline Totals

When nothing is reported at the top level, the dataset still ensures coverage by setting a baseline. For example:

• If (A+B) is missing, compute (A+B) = 100%.

This ensures that all company disclosures can be anchored in a complete taxonomy structure, even when top-level totals are absent. It preserves comparability across firms.

Takeaway

By applying these rules, Tracenable ensures that the EU Taxonomy dataset is:

• Complete – every KPI has totals and subcategories, even when disclosures are partial.

• Consistent – activity-level and total-level values always reconcile across eligibility categories.

• Comparable – disclosures from different companies and sectors are normalized into the same structure.

• Traceable – all computed values remain anchored in reported data, ensuring audit-grade reliability.

The result is a dataset that remains robust, transparent and decision-ready, even when company disclosures are incomplete or inconsistent.

---