METHODOLOGY / V2.0

Real checks. Clear limits.

A public website scan you can inspect, understand and repeat.

What we actually scan

One public URL, its initial HTTP/HTML response, robots.txt and one declared or default sitemap. No JavaScript rendering, authentication, full-site crawl, search-console data, ranking, traffic, backlinks, Core Web Vitals or AI visibility measurement.

Suprara fetches the URL you submit, follows up to three safe public redirects, checks robots.txt and requests one same-origin declared sitemap or /sitemap.xml. We respect robots restrictions for our scanner. If the page is blocked, inaccessible or not usable HTML, we show an error instead of inventing a score.

How the Score is calculated

Each evaluated check has equal weight within its dimension: pass = 1, fail = 0. Unverified and inapplicable checks are excluded. Each dimension is its pass rate ×100. The total is the average of evaluated dimension scores, rounded only at the end. At least three evaluated dimensions are needed for an overall score. Always read coverage beside the score.

These are Suprara-defined weights, not Google ranking factors. A high score with limited coverage does not prove site quality, indexation or visibility.
Identity & metadata

Page title

Meta description

Canonical declaration consistency

Indexability

HTTP page availability

Indexing directives

Robots crawling permission

Content structure

Text in initial HTML

Nonempty content headings

Main content landmark

Link discovery

Crawlable page links

Link labels

Discoverable sitemap

Accessibility signals

Declared page language

Image alternative attributes

No explicit zoom restriction

Structured data

JSON-LD syntax

Recognizable schema types

Structured URL values

Four honest result states

Pass: an observable check was met. Fail: an observable check was not met. Unverified: retrieval, rendering or parsing did not provide enough evidence; no penalty. Not applicable: a conditional or optional check does not apply; no penalty. Missing optional canonical, sitemap or JSON-LD is not automatically a failure.

Important limits

Only initial HTML is inspected. JavaScript-rendered content may not be visible to this scanner; that does not prove search engines cannot render it. Links are counted and labelled, not followed for broken-link testing. Structured data is checked for basic syntax and presence, not full schema compliance or rich-result eligibility. Accessibility checks are signals, not a compliance certification.

A report tied to its evidence

Your downloadable PDF uses the exact same saved scan, score, findings and timestamp shown in the application. It includes observed evidence, recommended actions, retrieved URLs, coverage and methodology. Rechecking requires publishing the changes on the actual website, then scanning the same URL again. Compare coverage as well as score.

Primary references

Scan a website