From Standards to Structure: Understanding the GRI Sustainability Taxonomy

As digital sustainability reporting becomes a regulatory necessity, the GRI Sustainability Taxonomy offers a practical solution for structured, comparable, and machine-readable disclosures.

GRI Taxonomy_1

On 19 June 2025, the Global Reporting Initiative (GRI) released the GRI Sustainability Taxonomy – a machine-readable digital representation of its Standards. Built on XBRL, the taxonomy enables organisations to prepare sustainability disclosures in structured formats, enhancing accessibility, accuracy, and interoperability.

Why the Taxonomy Matters

The push toward digital reporting is driven by regulatory momentum (e.g., CSRD, ISSB) and the growing need for robust sustainability data. GRI's taxonomy responds to this shift by offering an efficient way to generate, validate, and share sustainability information. Unlike previous efforts, the current landscape is technically ready – and GRI’s taxonomy is designed to meet that readiness.

What the Taxonomy Includes

The taxonomy fully reflects the GRI Standards structure: three Universal Standards (GRI 1–3), over 30 Topic Standards, and four Sector Standards (GRI 11–14). Key features include:

  • Dual submission options: automated (via XBRL) or manual (via webform);
  • Real-time validation to improve completeness and consistency;
  • Multi-format output: XBRL, iXBRL, Excel, and PDF;
  • Structured tagging aligned to the Standards down to subparagraph level.
  • The taxonomy was refined through targeted public consultations in 2024 with sustainability specialists and XBRL experts, ensuring it meets both technical expectations and practical reporting needs.

Illustrative Example_1

Source: Illustrative Example of XBRL Report

How It Is Built

The GRI Sustainability Taxonomy is distributed as a taxonomy package, structured in accordance with the Taxonomy Packages 1.0 specification. At the root level, the package includes a META-INF folder with the taxonomyPackage.xml, which defines metadata such as the taxonomy’s identifier, name, version, publisher, and publication date.

The core content is contained in the gri_srs folder, which includes schema files (.xsd) defining the elements of the taxonomy and their relationships. These include:

I. gri_srs_entry_point_2025-06-23.xsd – schema with ELRs for disclosures reported under GRI Standards.

II. core folder – contains schema files where GRI Standards elements are defined, including:

  • gri_srs_core_2025-06-23.xsd: contains all data points and properties;
  • gri_srs_part_2025-06-23.xsd: defines all typed dimensions used;
  • country-current-PR-2024-01-31: includes imported country taxonomy files;
  • lei: contains LEI taxonomy files and validation rules.

III. Formula folder – defines validation rules for disclosures:

  • formula_gri-2_2025-06-23: rules for GRI 2 disclosures;
  • formula_gri-3_2025-06-23: rules for GRI 3 disclosures;
  • formula_gri-101_2025-06-23: rules for GRI 101 series;
  • formula_gri-200_2025-06-23: rules for GRI 200 series;
  • formula_gri-300_2025-06-23: rules for GRI 300 series;
  • formula_gri-400_2025-06-23: rules for GRI 400 series;
  • formula_gri-generic_2025-06-23: general validation rules.

IV. labels folder – contains label linkbases:

  • lab_gri_srs_2025-06-23.xml: main English label linkbase file.

V. linkbases folder – includes modular presentation, definition, and reference linkbase files:

  • {pre\def}_GRI_disclosures_number_2025-06-23.xml: modular linkbase files for each Standard;
  • ref_gri_srs_2025-06-23.xml: reference linkbase file;
  • rol_gri_srs_2025-06-23.xsd: schema with ELRs for linkbases.

Label structures are mapped to three disclosure levels to mirror the structure of the GRI Standards. Level 1 refers to the main disclosure number (e.g., 2-1, 2-2), Level 2 reflects alphabetic sub-items (e.g., 2-3(a)), and Level 3 uses Roman numerals for further subdivisions (e.g., 2-3(a)(i)). This layered labelling system supports precise alignment between the human-readable and machine-readable versions of disclosures, ensuring consistency in interpretation and digital tagging.

GRI Taxonomy_Levels

Source: The GRI Sustainability Taxonomy Architecture and Implementation Guide

How It Works in Practice

The taxonomy supports tagging of disclosures using various data types:

  • booleanItemType is used for binary (yes/no) disclosures, typically beginning with the word “Whether…”. These elements indicate the presence or absence of a condition or action, such as whether financial statements are publicly filed or whether multiple entities are included in reporting.
  • enumerationItemType allows tagging based on a predefined list of options, such as Nature of Ownership, Legal Form, or Frequency of Sustainability Reporting. The taxonomy also includes enumerationSetItemType, which supports multiple selections for elements such as Countries of Operation or types of GHGs included in intensity ratios.
  • decimalItemType is used for numeric values such as monetary figures or intensity metrics, e.g., energy consumption or GHG emissions per unit.
  • stringItemType is applied to fields requiring unstructured or narrative inputs, including Legal Name, explanations, methodologies, and policy narratives.
  • dateItemType is used for disclosures that require date values, such as the start date of sustainability reporting or when a policy was first implemented.

Tagging is supported by real-time validation checks, including:

  • Derived mandatory rules to detect required disclosures;
  • Omission controls ensuring explanations are provided where applicable;
  • Non-negative validations for figures like GHG emissions;
  • Identifier checks for GRI Company ID fields.

Validation rules are assigned ERROR or WARNING severity. This supports assurance-readiness and helps organisations identify and correct gaps before submission.

Conclusion

The taxonomy is designed to balance flexibility and consistency. Optional typed dimensions allow for customised data breakdowns (such as region or measurement unit), while restrictions on extensions help maintain coherence across reports. Intensity metrics are further clarified through the use of tagged denominators (e.g., per m² or per FTE).

Reporting entities benefit from dual filing pathways – either through automated XBRL tools or a simplified web form. This adaptability makes the system accessible to organisations with varying technical capacities. GRI’s planned tools for third-party integration further support streamlined adoption.

The taxonomy enables multiple stakeholder groups to derive value: companies gain structured templates and quality checks; regulators can align policy design with reporting mechanics; analysts and assurance providers receive more consistent data; and digital tools, including those powered by AI, can reliably parse information for deeper insights.

Supporting Materials

The full GRI Sustainability Taxonomy package, along with technical specifications and illustrative examples, is available for download and includes:

GRI Sustainability Taxonomy (.zip)

GRI Sustainability Taxonomy (.xlsx)

GRI Sustainability Taxonomy Architecture and Implementation Guide (.pdf)

Illustrative Example of XBRL Report (.xml)