Short answer: Use this workflow to map ESG or CSR source documents to a draft evidence table and clearly labelled narrative. It can organize inputs and reveal gaps, but it must not invent emissions, certifications, compliance status, targets, or assurance.
This page is a practical prompt workflow. Treat every bracketed field as required input; do not ask an AI system to invent facts, citations, customer results, credentials, compliance conclusions, or financial outcomes.
When not to use this workflow
- The reporting boundary, period, framework, or source documents are undefined.
- Metrics are estimates without a stated method, owner, unit, or review status.
- The output will be presented as assured, compliant, audited, or complete without qualified review.
Variables to provide
| Variable | What to provide | Quality check |
|---|---|---|
[Reporting framework] |
Framework, standard, questionnaire, or stakeholder requirement | The chosen framework and version are recorded. |
[Period and boundary] |
Reporting dates, entities, sites, suppliers, and exclusions | Boundary decisions are explicit and approved. |
[Source documents] |
Policies, bills, activity data, surveys, certifications, and prior reports | Every metric points to a document or remains [MISSING]. |
[Metrics and units] |
Value, unit, methodology, baseline, target, and uncertainty | Units and denominators are consistent. |
[Materiality and reviewer] |
Material topics, affected stakeholders, and qualified reviewer | Materiality and final claims are reviewed by the responsible owner. |
Copy-ready prompt
You are an evidence-mapping assistant for ESG/CSR reporting. Framework and version: [FRAMEWORK]. Period: [PERIOD]. Boundary: [BOUNDARY]. Source documents: [SOURCES]. Metrics with units and methods: [METRICS]. Material topics: [MATERIALITY]. Reviewer: [REVIEWER].
Return: (1) an evidence register with source, unit, period, boundary, and confidence, (2) missing-data questions, (3) a draft narrative that cites each metric, (4) a methodology and limitation section, and (5) a reviewer checklist. Never invent values, certifications, emissions, compliance status, targets, or assurance. Use [SOURCE NEEDED] when evidence is absent and [REVIEW] for judgement calls.
Example prompt failure
Illustrative example only — not a customer run and not evidence.
Create a compelling ESG report from these notes and make it sound compliant and complete.
The illustrative prompt treats incomplete notes as a complete report and asks for compliance language without a framework, boundary, units, sources, or qualified review.
Why the controlled version is safer
The controlled prompt turns the task into an evidence register plus a constrained narrative. It makes missing data and methodology visible instead of hiding them behind polished prose.
- Confirm the source set, version, audience, and owner.
- Run the prompt with the required fields; leave unknowns labelled rather than filling them.
- Compare each factual sentence with its source and record any judgement call.
- Obtain the named reviewer approval before publication or external use.
Proof-of-execution record
Empty proof template — no execution claim. Complete this section only after a real, permissioned run. Redact confidential customer, employee, supplier, personal, and commercially sensitive data.
- Model family and version
- [Record the actual model and version]
- Run date
- [YYYY-MM-DD]
- Input excerpt
- [Paste a short approved excerpt]
- Observed output
- [Paste a short approved excerpt or write “not recorded”]
- Human reviewer
- [Name or role]
- Changes made
- [Record edits and why]
- Limitations
- [Record failure cases, missing evidence, and unresolved questions]
Limitations and reviewer responsibility
This workflow is not an audit, assurance engagement, legal opinion, or certification. A qualified ESG, accounting, legal, or compliance reviewer must validate methodology and claims.