User guide5 min

The site overview

Where a site opens: a score out of 100, what changed since the last scan, and a card for each audit area explaining the number.

The score#

A site with no failing checks scores 100.

Passing checks cost nothing. If your score is below 100, there is a specific list of reasons, and this screen shows all of them. There is no baseline deduction and no hidden component.

A site overview: the health score with its band and delta, how many checks broke and cleared since the last scan, the trend across runs, four counts, and a card for each audit discipline. — The score, what broke and what was fixed, the trend across scans, and a card per audit area below.

The band across the top, left to right.
Field	Example	What it does
Score	83 / 100	This scan's health score.
Band	Good	A word for the range the score falls in, so you do not have to interpret the number.
Delta	−1	The change since the previous scan of the same branch. A first scan says “First scan” instead of showing a zero.
broke / cleared	4 broke · 2 cleared	Checks that started failing, and checks that stopped, since the previous scan.
Trend	83-84 across 4 scans	A small chart of recent scores, with its range written out.
Counts	Pages · Open issues · AI readiness · Findings	Four figures, each linking to the screen it comes from.

How a failed check affects the score#

Severity sets the weight, and how many pages it affects scales it.

What one failed check costs before the number of affected pages is applied.
Field	Weight	What it does
Critical	10	Something that stops a page being found, indexed or trusted.
Important	5	Something that weakens a page without breaking it.
Standard	2	A small problem, worth fixing when you are editing the page anyway.
Suggestion	0	Advice only. It appears in the list but costs nothing, so a site can ignore every suggestion and still score 100.

That weight is then scaled by how many of your crawled pages the failure appears on. One mistake in a shared template affects every page and costs accordingly; the same mistake on a single page costs much less.

The score explainer dialog: the severity weights, the reach bands, and every problem with the points it is costing. — “How it's scored”, in the row of links below the score. It lists the weights and every problem making up the score, worst first.

The full formula is published

If you want the exact arithmetic, every weight and adjustment is written out. Read how the score is worked out

The row of links below the score#

Four shortcuts to related screens.

Field	Goes to	What it does
N scans	Scan history	Every scan this site has run.
Since the previous scan	Changes	What this scan changed compared with the last one.
How it's scored	A dialog	Every problem making up the score, with the points each one costs.
N unread alerts	Site alerts	Events raised for this site.

A card per audit area#

Trust, Content, Performance, Schema, Social and AI readiness, each summarising one Explore screen.

Each card shows a headline figure and the two or three worst problems behind it, and clicking it opens the full screen. Trust shows missing response headers and when your certificate expires; Content shows how many descriptions are flagged; Social shows how many cards are incomplete. A card with nothing to show says so rather than showing a zero.

The lower half of the site overview: where the work is by section, the split by kind of fix, and the checks grouped under config, template and per-page fixes. — Further down: where the work is, and whether each fix is a config change, a template change or a page-by-page job.

Do not compare scores between sites

The score is scaled by each site's own page count, so 83 on a 79-page marketing site and 83 on a 12,000-page catalogue do not mean the same thing. Compare a site with its own past scores instead.

If the product does not match this page, the page is wrong and we would like to know. Tell us