Companies may perform relevant activities without organizing the records in the form an assessment portal can evaluate.
Definition: activity and assessable evidence are different objects
A company may conduct relevant work yet be unable to demonstrate it in an assessment. A meeting, purchase or internal practice is difficult to evaluate when scope, ownership, date, deployment and outcome are not documented. Conversely, a polished document does not prove implementation. An evidence-shape problem exists when real activity has not been converted into attributable, current and reviewable records. It should be fixed through governance and recordkeeping, not by inventing certainty or reverse-engineering unsupported claims.
Operating model: create a controlled evidence library
Organize records by topic, entity, site, reporting period and evidence type. Distinguish policy commitments, endorsed procedures, implementation records, measured results and corrective actions. Each object should carry a title, owner, approval date, valid period, scope, version and source location. A policy without deployment evidence and a result without a method answer different questions. Reusable indexing can reduce duplicate work, but the same document should only support a response when its scope and date genuinely match the question.
Evidence model: build a requirement-to-record matrix
For each questionnaire item, record the relevant controlled documents, covered entity, period, evidence owner and confidence status. Add a short note explaining what the record proves and what it does not prove. Preserve the submitted version and assessment period so later teams can reproduce the answer. Where data are partial, state the covered locations or workforce rather than implying group-wide implementation. Where a measure changed, retain both versions and the reason. This produces an audit trail without predicting a particular rating outcome.
Failure conditions
Stop submission when a document is expired, unsigned where approval is required, outside the assessed entity, or cannot be connected to the answer. Do not upload marketing copy as proof of operating performance, a template as proof of implementation, or a target as proof of results. A high document count is not a quality metric. Duplicate, contradictory or excessively broad records can obscure the strongest evidence. If the team cannot explain an indicator's calculation or source, classify it as a gap and assign remediation rather than presenting it as verified.
Action checklist
Confirm the assessment scope and deadline. Export the question set into a responsibility matrix. Assign one evidence owner per topic. Inventory records before drafting answers. Check entity, date, approval, coverage and translation. Pair policies with implementation and result records where available. Remove duplicates and explain partial coverage. Run a reviewer challenge against every assertion. Freeze the submitted pack and keep a gap register for the next cycle. Never promise a score increase; measure progress through resolved evidence gaps and reproducible answers.
Primary source and scope boundary
EcoVadis describes its ratings methodology as assessing company sustainability management systems through policies, actions and results across environment, labor and human rights, ethics and sustainable procurement, with tailoring by factors such as industry, size and location. EcoVadis is a commercial assessment provider, not a regulator, and its current platform rules govern what evidence it accepts. This article offers preparation controls and does not predict a score. Official methodology overview: https://resources.ecovadis.com/ecovadis-solution-materials/ecovadis-ratings-methodology-overview-and-principles