Impact Score methodology

This document explains how each criterion is scored and combined. We use a mix of verified data, AI-assisted weighting, and statistical normalisation to keep the score consistent across product categories.

Scoring process

1. Collect structured sources (spec sheets, certification databases).
2. Apply criterion-specific scoring rules.
3. Weight the criteria based on relevance for the category.
4. Apply a data quality penalty if information is missing.

Statistical safeguards

We use sigma scoring to avoid over-valuing outliers. When the available data is too sparse, we fallback to percentile scoring so the ranking stays fair.

What this means for you

You can compare products inside the same category with a consistent scale, even when some attributes are incomplete.