"View source shows me my on-page tags."

"View source shows me my on-page tags."

Sometimes. And the times it doesn't are the ones that cost you.

The comparison: raw HTML (Ctrl+U, what the server sent) versus rendered DOM (what JavaScript built after). For a static or server-rendered page they're identical and view-source is fine. For a JS-heavy page they diverge — your title, canonical, or meta robots might be injected client-side and absent from the raw response.

Why it matters: Google renders, but on a delay and a budget. If your noindex or canonical only exists post-render, you're betting your indexing on the renderer queue.

So: use view-source to confirm what crawlers get on first pass. Use the rendered DOM (DevTools Elements, or URL Inspection's rendered HTML in GSC) to confirm what they get eventually. Check both, and panic specifically when a critical tag exists in one but not the other.

(The worst bug I see: canonical correct in rendered DOM, pointing at the wrong URL in raw HTML. Google read the raw one.)