Score Breakdown: the "How is this calculated?" Panel

Under every score on Security posture and on domain and run pages, a line summarises the score and a How is this calculated? button opens a side pane, "How this score is calculated". It works for three kinds of score: Security score, Standard rating and a single run's Rating.

What the pane shows, top to bottom

What it measures. One sentence for the kind of score you opened.

The areas we checked. "Each area is scored out of 100. 'Costs' is how many points it takes off the overall score." For every measured area:

• "Worth N% of the score".
• "Costs N points" or "Costs nothing".
• Its top three issues, or "Nothing found here lowered the score."

Not checked. "These are left out of the score, not counted as perfect, and not held against you." Each area is listed with the reason: no check has run, its check failed in the latest run, no data yet, or "Not in your plan, so it is not part of this score."

From areas to a grade. The score is the average of the checked areas weighted by how much each is worth. If the grade is provisional, a note says why. For the security score, a personalisation list shows how it "differs from your standard rating (N, grade) because of your own settings and decisions". Each line (an accepted risk, a changed weight, criticality, a module) has its own points, and together they add up exactly to the difference. A table lists the grade bands.

When and from what. Which runs the score was built from, with Open the run and links to each domain's page (up to eight shown).

Fine print. The methodology name and "A score is provisional when less than 60% of it could be checked."

Reading provisional grades

A provisional grade is a real grade on incomplete evidence. Do not report it as final. It stops being provisional when more of the score can be checked, for example after a broader rating run covers the areas listed under Not checked.

Use cases and wiring tips

• Audit trail: the pane names the methodology version and the runs, so a grade can be reproduced.
• Explain a gap between the scores to a colleague without opening Settings: read the personalisation lines.
• Find what to run next: anything under Not checked with "No check for this section has run yet" is fixed by running a rating that includes that check (Domain rating, run rating and Rate now).