This translation has not been editorially reviewed yet. The German version is authoritative. German version →
Performance
Core-Web-Vitals test: charging behavior with measured values
Charging behavior determines jump and visibility. We measure Core Web Vitals in the real browser — loading time, responsiveness, layout stability — per page of the scope of the audit, with metrics instead of estimates.
- Core Web Vitals (cwv-v1)Scope and rule set are stated in the report, countable.
- Human reviewClear decisions in the gold standard: a person judges disputed cases.
- TraceableEvery finding with reasoning, source and location.
- Publicly verifiableVia audit ID, QR code and an open verification page.
What is checked in detail
- Largest Contentful Paint (LCP): when the main content is visible
- Layout Stability (CLS): Does the page jump when loading?
- Responsiveness to inputs (INP-oriented laboratory measurement)
- Field data comparison where Google field data is available for the domain
- Findings per page — not just an average for the homepage
How your partial seal comes about
- Prove the domainDNS TXT record or file in the .well-known path.
- Choose the scopeBy page tier, up to 1,000 pages.
- Automated auditEverything technically decidable is measured.
- Auditor judgementA person judges disputed cases, in the gold standard.
- Partial sealFrom 90 of 100 points in this audit area.
The result: a verifiable partial seal
If your website reaches the target in this audit area, you receive a partial seal with its own audit ID and a public verification page with result, audit date and validity. How the audit proceeds is described in the audit procedure, the conditions under pricing.
Topics in detail
10 pages along the guidelines of the rule set, each with the associated checkpoints and its own questions.
- LCP prüfen: wann eine Seite für Besucher „da" ist
- FCP prüfen: wann das erste Pixel Inhalt erscheint
- INP prüfen: wie schnell eine Seite auf Eingaben reagiert
- Blockierzeit prüfen: wenn der Hauptthread keine Eingaben annimmt
- CLS prüfen: warum Inhalte beim Laden verspringen
- TTFB prüfen: wie lange der Server bis zum ersten Byte braucht
- Geschwindigkeitsindex prüfen: wie schnell sich der Bildschirm füllt
- Felddaten und Labordaten: warum zwei Werte für dieselbe Seite
- Mobil und Desktop getrennt messen: warum ein guter Wert nichts über den anderen sagt
- Keine Felddaten vorhanden: was dann gemessen wird
What a result looks like
Every criterion receives exactly one of three judgements. A criterion is never considered met just because the measurement found nothing.
- Met: demonstrably, with evidence
- Still open: awaits an auditor judgement
- Still to be met: location, rule and remedy named
A partial seal is issued from 90 of 100 points in this audit area.
Our own website currently reaches 100 of 100.
Good to know
- Measured in the real browser, not estimated.
- Values vary per network and device – this is why the report states the conditions.
