When the Corporate Sustainability Reporting Directive (CSRD) was announced, most companies focused on the content requirements: what data points to disclose, how to conduct a double materiality assessment, and which ESRS standards applied to their operations. Far less attention has been paid to a requirement that will fundamentally change how sustainability data is consumed, compared, and audited: mandatory digital tagging using XBRL.
In August 2024, EFRAG published the final XBRL Taxonomy for ESRS Set 1, providing XBRL tags for every data point and dimensional disaggregation defined in the disclosure requirements. This taxonomy covers over 1,200 data points across the full suite of ESRS standards. While digital tagging is not yet formally mandatory (ESMA must first develop the Regulatory Technical Standards and the European Commission must adopt them as a Delegated Act amending the ESEF Regulation), EFRAG has indicated that entities could be required to start digital reporting from the 2026 reporting year. By 2027, filing on the European Single Access Point (ESAP) will require XBRL digital tagging.
What Is XBRL/iXBRL Digital Tagging?
XBRL stands for eXtensible Business Reporting Language. It is a specialised extension of XML (eXtensible Markup Language) designed to make financial and non-financial data machine-readable and standardised. In practical terms, XBRL attaches a structured "tag" to each piece of data in a report, enabling computers to automatically identify, extract, compare, and analyse the data without human intervention.
Inline XBRL (iXBRL) is the specific format required under the European Single Electronic Format (ESEF). iXBRL embeds XBRL tags within an HTML document, producing a report that is simultaneously human-readable (viewable in a web browser as a normal document) and machine-readable (every tagged data point can be extracted and processed automatically). This is the format that CSRD sustainability statements will use: the report looks like a normal document to a human reader, but underneath, every data point is tagged with a standardised identifier that machines can process.
Why CSRD Requires Digital Tagging
Article 29d of the Accounting Directive, as amended by the CSRD, provides that undertakings shall prepare their management report in the electronic reporting format and shall mark up their sustainability reporting in accordance with the electronic reporting format. The rationale is straightforward: if 50,000 or more companies are going to produce detailed sustainability disclosures under ESRS, those disclosures need to be comparable, searchable, and auditable at scale.
Without digital tagging, a regulator, investor, or analyst wanting to compare the Scope 1 emissions of 500 companies would need to manually locate, extract, and normalise the data from 500 separate PDF or HTML reports. With digital tagging, a computer can extract every tagged data point from every report instantly, enabling automated comparison, benchmarking, trend analysis, and anomaly detection across the entire universe of CSRD reporters.
This has profound implications for data quality. When your sustainability data is digitally tagged and publicly accessible on the ESAP, errors, inconsistencies, and gaps become immediately visible to every stakeholder with access to an XBRL processing tool. The bar for data quality in sustainability reporting is about to rise dramatically.
What Needs to Be Tagged
The ESRS Set 1 XBRL Taxonomy published by EFRAG covers over 1,200 data points across all 12 ESRS standards. The taxonomy provides a specific XBRL element (tag) for every disclosure requirement and dimensional disaggregation. This includes quantitative data points such as Scope 1, 2, and 3 GHG emissions (in tCO2e), energy consumption by source, water withdrawal and discharge volumes, waste generation by type and destination, workforce headcount by gender, age, and region, and financial amounts related to climate risks and opportunities.
It also includes qualitative disclosures such as descriptions of transition plans, governance structures, risk management processes, and stakeholder engagement approaches. For qualitative disclosures, the XBRL tag identifies the disclosure requirement being addressed, enabling automated verification that all required disclosures have been included in the report.
The taxonomy distinguishes between mandatory ("shall") disclosures, voluntary ("may") disclosures, conditional disclosures (for example, those applicable only to companies in high climate impact sectors), and phase-in provisions that allow certain disclosures to be deferred for specified periods. Each XBRL element includes metadata indicating which category it falls into.
How AI Tools Automate Tagging
Manually tagging over 1,200 data points in a sustainability report is labour-intensive, error-prone, and impractical for most companies. AI-powered XBRL tagging tools automate this process by reading the sustainability report, identifying the content that corresponds to each ESRS disclosure requirement, and applying the appropriate XBRL tag from the EFRAG taxonomy. Natural language processing (NLP) models are particularly effective for tagging qualitative disclosures, where the tool must match narrative text to the correct disclosure requirement.
A list of certified XBRL software tools compliant with the specifications is maintained at software.xbrl.org. These tools validate that all mandatory tags have been applied, that tagged values are internally consistent (for example, that the sum of Scope 1, 2, and 3 emissions equals the total emissions figure), and that the dimensional disaggregations are complete and correctly structured. The XBRL International organisation provides the technical specifications and standards that underpin all compliant software.
Implications for Data Quality
Digital tagging fundamentally changes the data quality equation in sustainability reporting. In the current regime, sustainability reports are predominantly narrative documents where data is embedded in tables, charts, and prose. Errors are difficult for external parties to detect because there is no automated way to cross-check figures across different sections of the report or between reports of different companies.
With XBRL tagging, every quantitative data point is individually tagged and can be automatically validated against calculation rules embedded in the taxonomy. If your Scope 1 emissions by gas do not sum to your total Scope 1 figure, the validation will flag this immediately. If your energy consumption figures are inconsistent with your emissions calculations at standard emission factors, this discrepancy becomes detectable. The combination of digital tagging and mandatory third-party assurance under CSRD creates a data quality regime for sustainability reporting that approaches the rigour of financial reporting.
Preparing Your Data Architecture for Digital Reporting
Start with the taxonomy. Download the ESRS Set 1 XBRL Taxonomy from EFRAG's website. Map each taxonomy element to your existing data collection processes. Identify gaps where you are not currently collecting data at the granularity or in the format required by the taxonomy.
Design for machine-readability from the start. If you are building or upgrading your sustainability data management system, design it to store data in a structure that mirrors the XBRL taxonomy. This means storing every data point with the dimensional disaggregations required by the taxonomy (by facility, by scope, by gas, by sector, by region, by time period) rather than in aggregated form that must be disaggregated later.
Implement validation rules early. Embed the calculation linkbases from the XBRL taxonomy into your data collection process. This means that validation checks (such as ensuring that the sum of parts equals the total) are applied at the point of data entry, not at the point of report production. This prevents errors from propagating through the reporting process.
Plan for assurance. The combination of digital tagging and mandatory assurance means that your sustainability data will be subject to both automated validation (through XBRL processing tools) and human verification (through the assurance provider). Ensure your data management system produces a clear audit trail that links every tagged data point in the final report back to its source document, calculation methodology, and approval chain.
Conclusion
Digital tagging is not a minor technical formatting requirement. It represents a fundamental shift in how sustainability data is produced, consumed, and verified. When every data point in every CSRD report is tagged with a standardised XBRL identifier and publicly accessible on the ESAP, sustainability reporting moves from a document production exercise to a structured data management discipline. Companies that prepare for this shift now, by aligning their data architecture with the ESRS XBRL Taxonomy, implementing validation rules, and adopting AI-powered tagging tools, will be better positioned for compliance, produce higher-quality disclosures, and demonstrate the kind of data maturity that investors and regulators increasingly expect.
Share this post