Score Methodology

These four scores come straight from Google PageSpeed Insights, which runs Google's Lighthouse audit engine against your page using a mobile-device simulation. We don't invent the numbers; what you see is the official Lighthouse score (0-100) at the time the audit ran.

• Performance. How fast the page loads and becomes usable: main-content load time, server response, asset sizes, and render blockers. Measured in a lab simulation.
• SEO. Machine-checkable on-page basics: the page is crawlable, has a title and meta description, is mobile friendly, links are followable, and basic indexing status.
• Accessibility. How usable the page is for everyone, including screen-reader users: color contrast, image alt text, form labels, and heading structure.
• Best practices. Modern web standards and baseline security: HTTPS, no console errors, no deprecated APIs, and correct image handling.

The overall score at the top of the report is the simple average of those four categories, rounded.

When available, we use field data: the experience of real Chrome users on your site over roughly the last 28 days (the Chrome UX Report). This is the same data Google uses for page-experience assessment. If your site doesn't yet have enough traffic for field data, we use Lighthouse's lab measurement and label it as such.

The per-page Core Web Vitals chips in the crawl section are measured one by one via PageSpeed Insights for each crawled key page.

The issue list comes from our own crawler and analysis engine, plus Lighthouse performance audits. The principle: every finding must be evidence-backed. We only report what was actually observed, and every finding shows its evidence. Examples of the discipline we keep: a link is only called broken when the server answers 404/410 (not because our bot was blocked), and pages that are short by design, like a contact page, are never accused of thin content.

Server-level technical checks such as security headers, compression, and HTTP/2 support are tested directly against your server, including a real TLS negotiation for HTTP/2.

Organic-traffic and keyword figures come straight from your own Google Search Console (real, read-only data we never share). Backlink data needs a paid web-scale index and is not shown right now, so those fields stay empty rather than a fabricated number.

The AI Visibility analysis is a simulation on one real AI model: the model honestly judges whether it would mention your brand for five realistic user prompts. The result is directional, not absolute, and we label it as such.

Lab performance measurement always has natural variance: network conditions, your server's load at that moment, and Lighthouse version updates can shift numbers a few points between runs. Other tools may also measure with a desktop simulation (we use mobile, matching how Google evaluates) or a different Lighthouse version. What matters most is the trend over time and the status of each finding, not a one-or-two point difference.

Ready to see it live? Run a free audit or view a sample report.