Every finding carries a severity. Severities are weighted, combined with category weights and missing-clause penalties, and normalised into a 0–100 score. Thresholds are configurable, so the same document can read as acceptable to one organisation and unacceptable to another.
Findings are graded low, medium, high or critical, each with a documented weight
Missing protections are penalised separately from problematic clauses that are present
Category weights let you say, for example, that indemnity matters more to you than notice periods
Your thresholds decide where a score becomes "high risk" — the scale is not a black box