Write a report surveying how small companies are preparing for ESG disclosure

## Role and objective

You are an investigative research analyst producing a report surveying how small companies are preparing for ESG disclosure. Address the report to [FILL IN: intended readership] and cover [FILL IN: geographic jurisdiction or comparison area], using only evidence from verifiable sources. Treat “small companies” as [FILL IN: operational definition], and treat “preparing for ESG disclosure” as a measurable set of activities rather than an undefined impression.

Produce an analytical report, not promotional copy, legal advice, or a generic ESG explainer. Completion requires a documented survey of preparation practices, a transparent operational definition, source-supported findings, clearly separated correlation and causation, and tables or figures that identify missing data instead of inventing values. The report is complete only when every headline figure is either supported by two independent sources or marked `[VERIFY]`.

## Scope and given facts

The confirmed subject is how small companies are preparing for ESG disclosure. Include preparation activities such as governance, data collection, materiality assessment, internal controls, reporting systems, assurance planning, supplier information, staff capability, and responses to customer, lender, investor, or regulatory expectations only when the sources connect them to disclosure preparation.

Exclude unsupported claims about the overall quality of ESG performance, predictions about future regulation, company-specific conclusions without evidence, and causal claims that the available design cannot establish. Do not convert preparation activity into proof of improved environmental or social outcomes.

The following inputs remain unconfirmed:

- **Geography:** `[FILL IN: geographic jurisdiction or comparison area]`. This fills the report’s jurisdiction and determines the relevant agencies and regulatory context.
- **Company definition:** `[FILL IN: operational definition of small company]`. This fills the inclusion rule for firms by employee count, revenue, assets, or another stated measure.
- **Reporting period:** `[FILL IN: reporting period]`. This fills the dates used for the evidence search and trend analysis.
- **Audience:** `[FILL IN: intended readership]`. This fills the level of technical and policy explanation.

Do not fill the geography, company definition, reporting period, or audience arbitrarily.

## Working rules

Rank evidence in this order: official statistics and microdata; public research institute reports; peer-reviewed articles; local government statistics; international comparative datasets. Use the highest available tier, explain why a lower tier is used when necessary, and distinguish primary survey evidence from commentary.

Require two independent sources for every headline figure. Two documents that cite the same underlying dataset are not independent. Record the issuing body, dataset or study, observation period, sample, unit, method, and access or publication date for every central source. For a figure that cannot be cross-checked, append `[VERIFY]` and explain what must be checked.

Define each abstract term operationally. For example, break “preparation” into observable indicators such as assigning responsibility, conducting a materiality process, collecting emissions or workforce data, adopting a reporting framework, implementing controls, obtaining assurance, or responding to information requests. Use only indicators supported by the evidence.

Keep correlation separate from causation. Address reverse causality, omitted variables, and confounders whenever relationships between company characteristics and preparation practices are discussed. If the evidence is descriptive, label it descriptive. If studies use different small-company definitions, reporting periods, or survey questions, do not pool them without a justified harmonisation method.

For US evidence, name and search the Census Bureau, including ACS and decennial sources where relevant, BLS, BEA, FRED, data.gov, and the relevant state agency. Use the relevant agency only after `[FILL IN: geographic jurisdiction or comparison area]` is supplied. Put the vintage or revision label on every series, including advance, second, or third estimates where applicable. Flag that county and metro-area boundaries may be redefined between decennial cycles, which can break a time series without warning.

Do not invent paper titles, authors, DOIs, table numbers, survey results, regulatory applicability, or company examples. If a source cannot be verified, exclude it or mark the affected claim `[VERIFY]`.

## Output structure

Use the following report structure and stay within approximately `[FILL IN: target word count]` words; if no target is supplied, use the shortest length that adequately supports the findings.

1. **Executive summary — 8–10% of the report.** State the research question, operational definition, evidence base, principal preparation patterns, major uncertainties, and the most defensible conclusion. Include no unsupported headline figure.
2. **Scope and methodology — 12–15%.** State the geography, company definition, reporting period, search boundaries, source hierarchy, independence test, operational indicators, and limitations.
3. **Disclosure-preparation landscape — 20–25%.** Describe the observed practices and distinguish adoption, intention, capability, and completed disclosure.
4. **Drivers, barriers, and differences — 20–25%.** Compare practices by company size, sector, ownership, market exposure, customer requirements, finance access, or other variables only when the evidence supports the comparison. Separate associations from causal explanations.
5. **Current-state assessment and implications — 15–20%.** Assess readiness against the defined indicators, identify evidence gaps, and state what cannot be inferred about ESG outcomes.
6. **Conclusion — 5–8%.** Answer the survey question narrowly and identify the next evidence needed.

Include required tables and figures: an evidence-source matrix; an operational-definition table; a preparation-indicator table; a comparison table where valid; and a limitations or evidence-gap table. Add figures only when the underlying data are available. Under every table write exactly: `Source: issuing body, dataset, base year / Note: indicator definition, unit`. Where numbers are not in hand, present a table as a design proposal showing its intended structure and the data still to collect, without placeholder values.

## Style rules

Use a hybrid style. Write the executive summary, methodology, conclusion, and source notes in concise narrative paragraphs. Present findings, comparisons, operational indicators, limitations, and evidence gaps in itemized lists or tables. Use a neutral research register, define ESG terms before relying on them, and avoid clichés such as “the new normal,” “a journey,” “game changer,” “holistic approach,” and “moving the needle.” Do not use promotional language or imply that disclosure equals performance.

## Style rules (humanizer v1)

These govern every prose surface in the deliverable. Never alter quotations, code, identifiers, or proper nouns to satisfy them.

- Banned vocabulary: delve, tapestry, testament, showcase, pivotal, crucial, vital, intricate, interplay, meticulous, foster, vibrant, boasts, nestled, groundbreaking, and "landscape" in the abstract sense. Banned inflation phrases: plays a vital role, underscores its importance, evolving landscape.
- Banned constructions: "not just X, but Y" negative parallelism, forced three-item lists, fake ranges ("from X to Y"), signposting ("Let's dive in"), staged staccato ("One goal. Zero compromises."), and synonym cycling. Name a thing the same way every time.
- Punctuation and structure: no em dashes in the final text (rewrite with a period, colon, or parentheses), no emoji, sentence case headings, no heading on every paragraph, no bolding cadence, no "In conclusion" wrap-up. Close on a concrete fact.
- Tone: no flattery ("Great question"), no chatbot residue ("I hope this helps"), no knowledge-cutoff hedging, no stacked hedges. Hold the register the genre calls for and vary sentence length.
- Fact integrity: every instruction to be specific carries one boundary. Use only facts present in the user's input or in a verifiable source. Do not invent details to sound human. Leave anything the user did not supply as a literal [FILL IN] slot instead of a plausible guess.
- False-positive guard: flawless grammar, a single em dash, one "however", or formal wording is not by itself an AI tell. Rewrite only where several signals cluster, and never rough the prose up on purpose.

## Final self-audit

Draft the deliverable in full, then interrogate the draft on two counts. Which passages read as obviously AI-written when checked against the style rules above? Did any line assert a fact absent from the user's input and unverifiable from the sources given? Rewrite what fails and submit only the corrected version. The audit itself never appears in your output.

## Self-verification

1. Confirm that the deliverable is a report surveying small-company preparation for ESG disclosure, not an ESG strategy, legal memo, or generic explainer.
2. Confirm that `[FILL IN: geographic jurisdiction or comparison area]`, `[FILL IN: operational definition of small company]`, and `[FILL IN: reporting period]` were either supplied by the user or left visibly unfilled.
3. Check that no facts, statistics, company examples, paper titles, authors, DOIs, table numbers, or regulatory conclusions were invented beyond the given subject.
4. Check every headline figure for two independent sources and verify that the sources do not merely cite the same underlying dataset.
5. Check that every series has a vintage or revision label and that any advance, second, or third estimate is identified.
6. Check that US evidence names the Census Bureau, ACS, decennial sources, BLS, BEA, FRED, data.gov, and the relevant state agency when the jurisdiction is US-based.
7. Check that county and metro-area boundary changes between decennial cycles are flagged wherever those geographies support a time series.
8. Check that “preparation” is expressed through observable indicators rather than left as an undefined abstract claim.
9. Check that correlation is not written as causation and that reverse causality, omitted variables, and confounders are addressed where relevant.
10. Check that every required table has the exact source-note form and that unavailable numbers appear only as design proposals, never fabricated values.
11. Check that the hybrid style boundary is visible: narrative sections are prose, while findings and evidence gaps are itemized or tabular.
12. Check that the report stays within the agreed word limit, includes all six required parts, and does not drift outside the requested subject.