See what your website actually does
Paste a URL and get a measured report — not an opinion. Availability, TLS, technical SEO, performance, accessibility, UX, security, content and AI search readiness, each with the evidence behind it.
- No account needed
- About two minutes
- English and Spanish
- Free, and useful even if you never hire us
Audit a page
One address, one report. We fetch the public page once and measure what is there.
We fetch the page once, measure it, and keep the source only for a few days. Nothing is sold or shared.
What this installation measures
These numbers come from the audit rules stored in this installation, not from marketing copy. Administrators can enable, disable or re-weight any of them.
Nine categories, measured separately
Each category has a weight, and each check inside it passes, warns, fails or is not applicable. The total score is the weighted sum of what was actually measured — never a guess.
How the score is calculated
There is no black box. If a number appears in the report, you can trace it to the check that produced it.
- Every check returns pass, warn, fail, info or not applicable — with the measured value and where it came from.
- Categories are weighted, and the weights add up to 100. Checks marked not applicable are excluded from their category instead of counting as failures.
- Severity (critical, high, medium, low) is assigned per rule and shown next to every finding.
- We do not invent Lighthouse scores, Core Web Vitals field data, security exploits or AI rankings. If we cannot measure it, the report says so.
What each severity means
| Status | Description |
|---|---|
| Critical | Stops visitors or crawlers from using the page, or exposes something that should not be public. |
| High | A clear problem that costs traffic, conversions or trust, and should be fixed soon. |
| Medium | Worth fixing: it weakens the page but does not block anyone. |
| Low | A small improvement; fix it when the bigger items are done. |
| Information | An observation with no score attached. |
What you get at the end
A single report you can read, print or share, organised by category.
Score and grade
The overall result, the grade band it falls in, and the per-category breakdown that produced it.
Findings with evidence
Each finding shows what was measured, the value found, why it matters and what to do next.
Priority order
Critical and high items first, so a short amount of time goes to the changes that matter most.
Shareable link
A stable URL for this report, in either language, with no login required to read it.
AI search readiness
Answer engines and assistants read pages differently from people: they look for machine-readable structure, clear answers and permission to crawl. We check the things that are actually under your control.
To be clear: nobody outside those companies knows how ChatGPT, Gemini or Google rank a site, and we will not pretend otherwise. This category measures the technical conditions that make a page easy to read, cite and quote — nothing more.
Read the methodologyWhat gets checked
- Structured data: which schema.org types the page declares, and whether they parse.
- robots.txt and meta robots: whether AI crawlers are allowed, blocked or ignored by accident.
- llms.txt: whether the optional plain-text guide for language models is present and reachable.
- Sitemap: does it exist, is it valid XML, and are the URLs in it reachable.
- Answer-shaped content: headings, question patterns, lists and FAQ blocks an extractor can lift cleanly.
- Citation basics: canonical tags, hreflang, stable URLs and a clear site identity.
What this audit cannot tell you
Accessibility is automated only
Automated rules catch a share of WCAG 2.2 issues — missing labels, low contrast, empty headings — but they cannot judge meaning, reading order or keyboard flows. Manual testing with assistive technology is still recommended.
Security checks are passive
We look at public information only: headers, TLS settings, exposed files, cookie flags. No exploitation, no brute force, no port scanning, no injection attempts. A clean result here is not a penetration test.
Performance is measured, not predicted
Timings come from this server's single fetch of your page: DNS, connect, TLS, first byte and download. That is one observation from one location, not field data from real visitors and not a Lighthouse run.
Questions people actually ask
Is the audit really free?
Yes. There is no account, no card and no trial timer. The report is meant to be useful whether or not you ever work with the studio.
Do you store my website?
We fetch the page once to measure it. The HTML source may be cached for a few days so the report can quote evidence, then it is deleted automatically. We do not crawl the rest of the site unless the audit asks for a small number of internal pages, and the limit is configurable.
How is this different from Lighthouse or PageSpeed Insights?
Those tools are excellent and we do not replicate them. This audit is wider: it combines availability, TLS, technical SEO, performance signals, accessibility, UX, security headers, content and AI search readiness into one weighted score, and every finding explains why it matters and what to do. We never pretend to have run Lighthouse.
Can I audit any website?
Any public http or https page. Addresses that resolve to localhost, private ranges, link-local or cloud metadata endpoints are refused before any request is made, so the tool cannot be pointed at internal infrastructure.
Why is a check marked "not applicable"?
Because the condition it tests does not exist on that page — a page with no images has no image alt text to check, a page with no forms has nothing to label. Not applicable checks are excluded from the category score instead of being counted as passes or failures.
Who built this, and what happens to my details if I ask for help?
Qvicom Marketing & Design Studio, a bilingual studio in Miami and Bogotá. If you send the report form with your contact details, we store them to reply to you, and only for that. Consent is never pre-selected, and you can ask for deletion at any time.
Fixing it is the other half
A report is only worth something if the items in it get done. Run the audit first — if you would rather have someone else implement it, the studio that built this tool does that work too.
No pressure, no countdown, no pop-up. The audit is the product; the studio is optional.