EcoVadis rating
The medal is won in the methodology.
Most teams spend two months filling in the questionnaire. Palau reads the scoring methodology so you don't have to — drafting every answer from evidence you already have, predicting the score per question, and prioritizing what actually moves your rating.
Built on your previous submission, your scorecard feedback — and the rules EcoVadis scores you by.
The problem
EcoVadis isn't a questionnaire. It's a scoring system — with coverage multipliers, weighted layers and medal thresholds that climb every year. Most teams answer the questions without ever reading the rules they're scored by.
So the same things keep happening: two or three months of work for one or two people, documents rejected on format technicalities, points lost on options nobody knew were supported by evidence that was already in the annual report — and next year, it all starts from scratch. The rating drifts down while the work stays the same.
Environment · Policies
Environmental policy covering group operations
75
predicted score
- Quantified targets, group scope — supportedEnvironmental policy · p. 4
- Quick win: option you didn't select last year is now supportedSustainability report · p. 31
- ISO 14001 certificate expires before submission — renewal flagged
- To reach 100: external recognition of the framework — action drafted
Every question drafted, scored and prioritized — before you open the portal.
How it works
From scorecard to submission, in four moves.
Your previous submission is the most valuable document you own. Palau starts there — not from a blank form.
Load last year
Your scorecard, your previous answers, the analyst feedback, every rejected document and why. When EcoVadis changes the questionnaire — new mandatory questions, tightened rules — your workflow is rebuilt with the delta applied.
Drop in your evidence
Policies, reports, certificates, HR metrics — into the Vault, where AI categorises each file and tracks validity. The same evidence base that feeds your ESRS report feeds this.
AI drafts and scores
One pass drafts every answer with document and page named per option. A second pass predicts the score, question by question, against the actual EcoVadis methodology — and says exactly what's missing for the next increment.
You review, we submit
Accept, edit or reject each draft. Fill the gaps the AI flagged. Then export back to the portal — or hand us the whole thing, from chasing colleagues to uploading the final submission.
What the workflow does
Built by people who've read every page of the methodology.
Not a text generator pointed at a form. Six things that decide whether your score goes up.
Every answer, page-referenced
For each supportable option: the exact document, the exact page, and a short analyst comment on why it qualifies. You verify and copy — you don't hunt.
A predicted score per question
The 0–25–50–75–100 rubric applied before you submit, with what's missing to reach the next increment — weighted by how much each question actually contributes to your theme.
Quick wins you already earned
Options you didn't select last year that your current evidence now supports. Points sitting in documents you already have — surfaced before anything new gets written.
Evidence forensics
Stale KPIs, expired certificates, wrong scope, documents that will bounce on format — flagged before submission, not after. Rejected documents from last year are checked for proper replacement.
Medal math, built in
Coverage multipliers, compensation bonuses, layer weights — the structural rules that make or break a rating are modeled into the workflow, so effort goes where the points are.
Backup when the analyst is wrong
Rejections are cross-checked against your evidence and the methodology. When the assessment itself has an error, you get the argumentation to challenge it — inside the appeal window.
50%+
less manual effort — a minimum we've committed to in client contracts
58
questions in a full EcoVadis questionnaire, drafted in a single run
0–100
the full scoring rubric modeled — increments, weights, multipliers
2026
questionnaire delta already applied — new mandatory questions included
The part nobody tells you
Why good answers still lose points.
Every theme is scored in three layers — and they don't behave the way you'd expect. Coverage doesn't add points; it multiplies them. A strong set of actions with thin site-level coverage collapses to a weak Actions score, and the Actions layer is the biggest one there is.
Meanwhile, medals are awarded by percentile, not absolute score — the thresholds climb every year as the rated population improves. Standing still means sliding down. And in Ethics and Sustainable Procurement, a compensation bonus can substitute for certifications that don't exist in your industry — if you know to trigger it.
These rules are public. Almost nobody reads them. We turned them into software — so every draft, every predicted score and every priority flag already accounts for how the points really flow.
- Policies25% of theme
Policies & endorsements — the paper trail.
- Actions40% of theme
× coverage multiplierMeasures & certifications — thin site coverage scales the whole layer down.
- Results35% of theme
Reporting & 360° Watch — what you publish, and what the media says.
Medals awarded by percentile within your industry and size peer group — thresholds drift upward every year. Compensation bonus modeled for Ethics & Sustainable Procurement.
Beyond the submission
One evidence base. Every rating that hits your inbox.
CDP, tenders, customer questionnaires
Sustainability teams field a hundred-plus questionnaires a year. The same vault that answers EcoVadis answers the rest — with evidence attached.
Multi-country, multi-entity programs
Group scorecard plus country submissions, each built from the same governed data — one client runs EcoVadis across seven countries this way, on local reports built in the same platform.
Fully managed, if you want it
For teams with no bandwidth at all: we run the whole cycle — stakeholder chasing, evidence collection, drafting, review and the final portal submission. End to end.
ESRS and EcoVadis, same twin
If you're already reporting under ESRS on Palau, your EcoVadis submission starts 80% done — the evidence, the entities and the narratives are already governed.
See the ESRS workflow
The AI agent doing the reading
Reef is Palau's AI agent — not a chat window. For EcoVadis it has read the full scoring methodology, your previous submission, the analyst's feedback and every document in your vault. That's why it can name the page, predict the score and rank the fixes.
It's the same agent your peers ask for gap analyses, dashboards and benchmark reports — trained on sustainability frameworks, not on the open internet. Ask it which of three documents makes the strongest evidence for a question, and it will tell you — with reasons.
See it run on your scorecardWhat we won't pretend to do
360° Watch is EcoVadis's own media monitoring — nobody controls it, including us. The carbon module runs on EcoVadis's side; we ingest the results. And no tool can turn missing practice into points: when the evidence genuinely isn't there, Palau tells you what to build, not what to claim.
“It really takes quite a while to complete EcoVadis and it's only me and one colleague. This will already really help save us time — I really appreciate what it looks like and the help that you're offering.”
Sustainability lead, global materials group — during their first EcoVadis working session on Palau
See who submits through PalauEcoVadis® is a registered trademark of EcoVadis SAS. Palau is an independent platform and is not affiliated with, endorsed by, or sponsored by EcoVadis.
Bring last year's scorecard. Leave with your medal path.
In a pilot, we load your previous submission and scorecard feedback, run the AI across your evidence, and walk you through it question by question — where the points are, what's a quick win, and what needs real work. Before you see a contract.