Evidence-based · no fabricated metrics
Inspect your website’s SEO. See the evidence behind each finding.
Enter a public URL to inspect its technical SEO and see the evidence behind every finding.
- 100+
- criteria with explicit evidence or limits
- 4 + 4
- primary readiness and supporting evidence categories
- Clear scope
- measured, heuristic and unverified findings
Returning with a backup?
Restore saved audit summaries and comparison snapshots locally. The file never leaves this browser.
No local-data restore action has run.
Sample report
See the report before you run one
Screenshots of a real saved audit of this site, not mock-ups. Each finding shows the value it was read from and the source behind the rule.
Free, no sign-up. Saved reports stay in your browser, and every criterion is tied to one of 89 published primary sources.
Coverage
What gets analyzed
Four primary categories determine search-implementation readiness when their evidence is sufficient. Four supporting categories add useful diagnostics without changing that central number. Every category keeps its own measured scope, and unavailable checks are shown rather than estimated.
- Readiness input
Crawlability & Indexing
Whether search engines can reach, fetch, and index this URL at all — HTTP status, robots directives, robots.txt and sitemap discovery.
- Readiness input
Technical Foundations
HTTPS, redirect behaviour, canonical correctness, language declarations, hreflang, viewport, character encoding, mixed content and DOM complexity.
- Readiness input
On-Page Signals
Title link, meta description snippet, heading outline, image alt semantics, and internal/external link quality.
- Supporting
Content Signals
Observable content signals: text structure, readability, repetition, freshness markers and contact links. These are limited proxies; they do not establish usefulness, expertise, accuracy or whether a page satisfies its purpose.
- Readiness input
Performance
Server-measured transfer characteristics, Core Web Vitals from real-user field data, and a separately identified Lighthouse lab score when available.
- Supporting
Structured Data
JSON-LD, Microdata and RDFa discovery, syntax validation, Schema.org type recognition, and required/recommended property checks for Google-supported features.
- Supporting
Static accessibility markup
Static, HTML-verifiable accessibility signals: document language, alternative text, form labels, accessible names for controls and links, landmarks and iframe titles.
- Supporting
Social Metadata
Open Graph and X (Twitter) card metadata, favicon, site identity and discoverable social profiles.
Checking a specific file or URL?
Use a focused tool to inspect it and review issues before publishing.
Process
How it works
Start with the entry page, then review its findings and the coverage of any additional pages inspected.
Choose your scope
Check one public page, or choose a deep audit to inspect a sample of up to one hundred pages on the same site.
Inspect the HTML
We read the HTML the site serves: titles, metadata, headings, images, links and structured data. To review script changes, compare safely pasted rendered HTML or use an isolated renderer when the deployment provides one.
Evaluate criteria
Each criterion explains its status and evidence. Checks that need unavailable data remain unverified. Performance providers are queried separately.
Score and prioritise
Review the measured findings, their suggested fixes and the inspection coverage. Missing measurements do not count as passes or failures.
Read the score with its coverage
The score summarizes the criteria measured in this audit. It is not a Google ranking prediction or proof that unchecked pages have no issues. Check the evidence and any crawl or performance limits before deciding what to change. See how the checks and scoring work.
For developers
From finding to clean, reviewable code
A useful audit should finish inside the codebase. SEOLens keeps the evidence, implementation decision and verification step connected so a developer can review the change with confidence.
Read the exact evidence
Start with the observed value and the HTML, header or provider that produced it.
Apply a confirmed fix
Change only the signal you have verified. The example keeps three URL declarations consistent.
Re-audit and compare
Run the page again and confirm the evidence changed before treating the issue as resolved.
// Align every URL signal to the confirmed preferred page.
import type { Metadata } from 'next';
const preferredUrl = 'https://example.com/services';
export const metadata: Metadata = {
alternates: { canonical: preferredUrl },
robots: { index: true, follow: true },
openGraph: { url: preferredUrl },
};Illustrative implementation after confirming /services as the preferred URL. Use index: true only if this page is intended to appear in search. The audit provides evidence; the developer keeps the product decision.
Principles
How this differs
Plenty of tools will hand you a confident number. These are the constraints this one holds itself to instead.
No invented data
Measurements come from the fetched page, crawl, or a named performance provider. What cannot be measured says “unable to verify” and is excluded from the score rather than estimated.
Transparent arithmetic
Every report shows its own arithmetic: points earned, weights, and each contribution summing to the final score.
Primary sources only
Criteria come from Google Search Central, web.dev, Schema.org, WCAG and the RFCs — never third-party commentary.
Honest about heuristics
Google publishes no title length limit. Thresholds like that are marked as heuristics, never presented as requirements.
Questions
Frequently asked questions
What does SEOLens Evidence check?
More than one hundred criteria across eight evidence categories. Crawlability, technical foundations, on-page signals and performance feed the central search-implementation readiness model when enough evidence is available. Content, structured data, static accessibility markup and social metadata stay visible as supporting diagnostics. Each criterion returns a status, its evidence and a primary source.
Is SEOLens Evidence free to use?
Yes. There is no account, no sign-up and no cost. Rate limits apply per IP address so a single client cannot exhaust the analysis capacity for everyone else.
Does it measure Core Web Vitals?
After the HTML audit, available Chrome UX Report field data can appear while a separate Lighthouse lab test runs. Core Web Vitals checks use 75th-percentile real-user measurements and identify whether they cover the URL or its origin. A missing field metric is “unable to verify” and carries zero weight. Recent performance results may be reused for up to five minutes, with timestamps and a refresh option. Lighthouse describes a simulated load and does not substitute for real-user Core Web Vitals.
Can it audit a whole site rather than one page?
A deep audit can crawl up to one hundred pages of the same site, subject to the selected page limit and time budget. It obeys robots.txt and the crawl-delay it declares, then reports patterns in the pages it inspected alongside the entry page’s own findings. It identifies any pages or checks it could not complete.
Which pages can be analyzed?
Any publicly reachable HTTP or HTTPS page. Pages behind a login, private networks and loopback addresses are rejected by the URL policy, because the analyzer fetches server-side and must not be usable to reach internal hosts.
Who built SEOLens Evidence?
SEOLens Evidence was developed by Razan Aboushi, a software engineer working on search, performance and large-scale web platforms. Every criterion it applies is documented in full on the methodology page, including the thresholds and what the tool cannot determine.
Get started
Ready to check a page?
Return to the analyzer above, enter a public URL, and choose a quick page check or a deeper site crawl.
Go to the website analyzer