What the Free Site Audit Actually Checks
The free site audit runs eight concrete checks against a homepage's raw HTML — the response a crawler sees before any JavaScript runs. None of them are exotic. They're the fundamentals that quietly break, and each one is verifiable in your own page source. Here is exactly what the tool looks for and why each check earns its place.
1. Title tag
One <title>, present and specific. It's the strongest single line
of on-page metadata: it names the page in search results, browser tabs, and link
previews. The audit flags a missing, empty, or duplicated title element.
2. Meta description
Not a ranking factor, but it's your search-result pitch — the difference between a listing that gets clicked and one that gets skimmed past. The audit checks it exists and isn't boilerplate.
3. Open Graph tags
og:title, og:description, og:image,
og:url — what your page becomes when someone pastes it into LinkedIn,
Slack, or iMessage. Missing OG tags means link previews fall back to whatever the
platform scrapes, which is usually wrong.
4. Twitter Card tags
Same idea, X/Twitter's dialect. The audit verifies a card type is declared so shares render as rich cards rather than bare links.
5. Viewport
Without <meta name="viewport">, phones render the desktop layout
zoomed out — and search engines evaluate the mobile experience first.
6. Canonical
The rel="canonical" link tells search engines which URL is the real one
when several paths serve the same content (trailing slashes, query strings, www/non-www).
Absent or wrong canonicals split ranking signal across duplicates.
7. Robots directives
The audit reads what your page is telling crawlers. An accidental
noindex ships to production more often than anyone admits, and it's
invisible until traffic disappears.
8. Charset
A declared <meta charset> prevents encoding ambiguity — mundane
until a curly quote renders as mojibake in a search snippet.
Why raw HTML, not the rendered page
Every check runs against the raw HTTP response. JavaScript can inject any of these tags after load, and Google can often (not always) render them — but the raw response is what every crawler, preview bot, and AI answer engine sees first, and it's the only version you fully control. If the fundamentals are right in the raw bytes, nothing downstream can un-break them.