How are suppliers tracking?

Used the moment data is uploaded -- no waiting, no API cost. It always starts from whatever Score/Rating/Priority/Disclosure Completion your file already has, and only estimates what's missing:

This is a transparent estimate, not measured E/S/G data -- your files have one overall Score per supplier, not separate E/S/G indicators. It's clearly marked as calculated/estimated everywhere it appears, and is superseded automatically by real AI-verified figures below.

Triggered from "Verify with AI research" (in a supplier's drawer, or "Verify all with AI research" in Impact Profiles). This calls a live AI model with web search to research the supplier and produce a real E/S/G breakdown, Confidence score and cited sources -- overriding the baseline estimate above wherever it appears. This is the exact prompt sent every time; edit it below and your changes take effect on the next run. {{NAME}} and {{CONTEXT}} are replaced automatically with the supplier's name and known context (country, sector, spend, emissions). The JSON output shape at the end must stay intact, or results can't be parsed.

The Klappir Impact Profile™ is a standardized supplier sustainability assessment engine that evaluates suppliers based on Environmental, Social and Governance (ESG) performance, disclosure quality and procurement risk. It provides a consistent, transparent and auditable supplier assessment that supports:

The assessment produces a numerical score (0-100), an ESG Rating (A1-D), a Confidence Score and a Priority classification.

AI Research Layer

The AI research engine collects and summarizes publicly available information including company overview, sector, environmental impacts, social risks, governance practices, sustainability commitments and public disclosures. Every statement includes a reference to the original source.

Each supplier receives an ESG score from 0-100, made up of three weighted categories:

Measures the completeness and reliability of publicly available information. Calculated independently from the ESG Score (maximum 100%). Indicators include Sustainability/Annual/ESG/Climate Report, Scope 1-3 disclosures, Policies, Certifications, SBTi, CDP, EcoVadis, External Assurance. Example: 85% = high confidence, 40% = limited confidence. The Confidence Score must never influence the ESG Score directly -- it indicates the reliability of the assessment, not the score itself.

Convert the numerical score into a rating using these thresholds:

Calculated independently from the ESG Rating. Considers Annual Spend, Estimated Scope 3 emissions, Strategic importance, Country Risk, Sector Risk and Confidence Score. Priority values: Critical High Medium Low.

For every supplier, generate:

Each supplier profile shall contain: Supplier, Website, Telephone, Email, Country, Sector, Company Description, Value Chain Role, Environmental/Social/Governance/Economic Summary, Disclosure Completion, Available Disclosures, Environmental/Social/Governance Score, Overall Score, Rating, Confidence Score, Priority, Sources, Last Updated.

Display: Overall Score (large numeric), Rating Badge (A1-D), Confidence Gauge (0-100%), Priority Badge (Critical/High/Medium/Low), Radar Chart (Environmental/Social/Governance), Disclosure Completion, Timeline (score history), Comparison (industry average, country average, top suppliers).

The architecture shall support future AI functionality including:

The assessment engine is modular so new indicators and scoring criteria can be added without changing the core architecture.