ShipGuardeRELEASE CLEARANCE BUREAU
How it worksUse casesDocsPricing
FREE CHECK · PERFORMANCE

Core Web Vitals, with the element named.

Paste a public page. It loads it in a real browser, measures Largest Contentful Paint, Cumulative Layout Shift and Total Blocking Time, and tells you which element was responsible.

All six checks
RUN IT ON YOUR SITEFREE · NO ACCOUNT

Six read-only checks on one page: accessibility, performance, SEO, console and network errors, broken links, cookie handling. About a minute. Nothing is submitted, signed into, or changed.

WHAT IT LOOKS AT

The questions it asks of the page.

01Largest Contentful Paint, and which element it was
02Cumulative Layout Shift, and what moved
03Total Blocking Time from long tasks on the main thread
04A 0 to 100 score on the same log-normal curve Lighthouse uses, so the number is comparable
05Whether the page actually loaded: a 403 or an empty shell is reported as unmeasured, not as a perfect score
06Render-blocking resources and oversized images on the critical path
WHAT IT FINDS MOST OFTEN

From real public pages. No company named.

FOUND
LCP OVER EIGHT SECONDS ON A MARKETING HOMEPAGE

A hero video autoplayed before the headline painted. Visitors on mobile saw a blank page for most of ten seconds. Poster image first, video after the headline is visible.

FOUND
LAYOUT SHIFT OF 1.9 FROM A COOKIE BANNER

The consent banner injected itself above the content after load and pushed the whole page down. Reserve the banner's space, or overlay it instead of inserting it into the flow.

FOUND
A PERFORMANCE SCORE OF 100 ON A PAGE THAT NEVER LOADED

This was our own bug: a bot wall returned 403 and the scorer measured an empty document. Fixed, and now reported as unmeasured. If you see "unmeasured", the page blocked the check. Allowlist the checker or test a page that is not behind a wall.

WHAT IT CANNOT SEE

The honest limit.

One load from one location on one device profile. It is a measurement, not a field-data average, and it will differ from what your users in another region see. Treat the element it names as the lead, not the score.

Run the journey too: the pilotCheck a page now
INQUIRIES

Questions about the performance check.

How is the score calculated?
Each metric is scored on the log-normal curve Lighthouse publishes, then combined with the same weights. Earlier versions bucketed metrics into three bands, which meant a page at 4 seconds and a page at 8 seconds scored the same; the curve fixed that.
Why is my number different from PageSpeed Insights?
Different location, different device profile, different moment. Lab numbers move. The element the check names, and the reason, are the stable part.
What does "unmeasured" mean?
The page did not load well enough to measure: a 403, a redirect loop, or an empty document. The check refuses to score what it did not see.
THE OTHER FIVE

Every run does all six.

01Accessibility
02Broken links
03SEO basics
04Console & network errors
05Cookies & consent

Code review checks the diff. ShipGuarde also checks the running product, so a release does not ship on the hope that nothing broke.

Join the free 30-day pilotStart a 7-day trial

NO CREDIT CARD · CONNECT GITHUB IN UNDER FIVE MINUTES · CANCEL ANY TIME

ShipGuarde

RELEASE CLEARANCE BUREAU
ISSUING AUTHORITY FOR SOFTWARE RELEASES
[email protected]

PRODUCT

  • Features
  • PR Reviews
  • Visual QA
  • Compare
  • Integrations
  • Pricing

RESOURCES

  • Free website check
  • Docs
  • How it works
  • Use cases
  • Sample report
  • Founding pilot
  • Blog
  • Changelog
  • FAQ
  • Support

COMPANY

  • About
  • Security
  • Status
  • Contact us
  • Privacy
  • Terms
  • Refunds
  • Sub-processors
FILED · SG-482 · ACME-COMMERCEEST. 2026