Scope
We assess Fortune 500-scale companies using consistently available public evidence.
How we work
A transparent research framework for measuring the decisions behind corporate responsibility.
We assess Fortune 500-scale companies using consistently available public evidence.
More than 2,000 articles and 2,000 Fortune 500 CSR data points are reviewed against documented research standards.
Pillar-level outcomes are normalized and weighted into an interpretable overall score.
A five-person research team reviews material findings, sources, and limitations.
Comparable indicators, outcome quality, disclosure, and external validation.
Comparable indicators, outcome quality, disclosure, and external validation.
Comparable indicators, outcome quality, disclosure, and external validation.
Comparable indicators, outcome quality, disclosure, and external validation.
Scores are research tools, not investment advice. Every model has limitations; our methodology documents them plainly.
Data documentation
Researchers preserve the source, reporting year, access date, measurement boundary, and verification status so every published observation can be traced back to evidence.
| Documentation | Included evidence |
|---|---|
| Data sources | Annual reports, sustainability and CSR reports, SEC filings, governance documents, company disclosures, regulator records, and credible independent reporting. |
| Variables | Company, industry, reporting period, indicator, unit, value, baseline, target, outcome, source URL, assurance status, and analyst confidence. |
| Evidence treatment | Missing evidence is separated from negative evidence; modeled values and unverified claims remain visibly labeled. |
Qualitative risk methodology
Each industry maps only to material categories supported by government or regulatory research. A category is an exposure lens, not evidence of an incident.
The related CSR pillar is compared with the live industry average. Ten or more points below peers is labeled High; four to under ten below is Elevated. A Moderate category may be Low when it is ten or more points above peers.
The overall qualitative profile equals the highest displayed category. It is not a numerical risk score, prediction, legal conclusion, or automatic CSR-score deduction.
Standardized interpretation
Overall scores translate to one of five letter grades. Decimal scores use the same boundaries: a score of 89.9 is a B, while 90.0 is an A.
90–100
Excellent
80–89
Strong
70–79
Developing
60–69
Needs improvement
Below 60
Insufficient
Weighting model
Equal pillar weights make the model legible. Indicator-level treatment still accounts for evidence strength, reporting continuity, and sector materiality.
| Pillar | Weight | Research scope |
|---|---|---|
| Environmental | 25% | Climate, resource use, operations, and measurable outcomes |
| Financial responsibility | 25% | Long-term resilience, stakeholder value, and responsible allocation |
| Philanthropy | 25% | Community investment, participation, and documented public benefit |
| Ethics | 25% | Governance, conduct, accountability, and external validation |
Overall score = Σ (pillar score × 0.25)
Scores are rounded to one decimal place after aggregation.
Pillar rubric
Each pillar uses the same evidence-quality questions while allowing material indicators to reflect the operating realities of the company’s industry.
| Rubric dimension | Research test |
|---|---|
| Commitment | Is there a defined policy, baseline, scope, and time-bound objective? |
| Governance | Is ownership assigned and does oversight reach executives or the board? |
| Implementation | Are resources, controls, programs, and operating changes documented? |
| Outcome | Are comparable results reported over time against the stated objective? |
| Verification | Is the evidence current, traceable, and independently assured where appropriate? |
Research workflow
Questions and limitations
The full index is published annually. Material corrections and source updates are reviewed throughout the research cycle.
Companies may submit public evidence or corrections, but cannot purchase placement, suppress findings, or influence model weights.
Missing evidence is distinguished from negative evidence. Confidence and disclosure quality are evaluated separately before normalization.
No. Impact500 scores are comparative research tools and are not investment, legal, or compliance advice.
Limitations
Public evidence is incomplete and uneven. Scores describe documented performance within a defined model, not every dimension of corporate conduct.
Research principles
Every conclusion must be traceable to evidence, consistently evaluated, and open to revision when stronger information appears.
Version history
The current release strengthens outcome validation, source freshness, AI governance, and cross-sector normalization.