SEO Vibes

Checklist

SEO Checklist for AI-Built Websites

Not a flat pile of a hundred items. Twelve layers in dependency order, because on an AI-built site the failures cluster — and fixing the wrong layer first wastes the effort.

By SEO Vibes11 min read

Most SEO checklists are a flat pile of a hundred items with no order and no theory of what depends on what. That is useless for an AI-built site, where the failures cluster in a few specific layers and where fixing the wrong layer first wastes the effort. This one is organized into twelve layers, in dependency order — earlier layers gate later ones, so work top to bottom.

The rule underneath the order: access before meaning, meaning before competition.A page that can’t be crawled can’t be indexed; a page that isn’t indexed can’t rank; and content, authority, and conversion only matter once the first three are solved.

The twelve layers

What each layer answers

Read this as a map. The deep checklist for each layer follows — but the order here is the order to work in.

1 · Technical access
Access
Can a crawler reach the page and get a normal response?
2 · Rendering
Access
Is your real content in the HTML, or only after JavaScript runs?
3 · Indexation
Access
Are you telling Google to index the page — and the right version of it?
4 · Architecture
Access
Do URLs and site structure make the important pages reachable?
5 · Metadata
Meaning
Does each page describe itself with a unique title and description?
6 · Content
Meaning
Is the content substantial, specific, and written for a real reader?
7 · Internal links
Meaning
Do real links connect pages so Google can discover and weigh them?
8 · Structured data
Meaning
Have you described entities in a machine-readable way where it helps?
9 · Performance
Competition
Does the page load and become stable fast enough not to be a liability?
10 · Authority
Competition
Do other credible sites reference yours at all?
11 · Conversion
Competition
Once found, does the page turn a visitor into a customer?
12 · Measurement
Competition
Can you see what is indexed, what ranks, and what converts?

Layers 1–4 are where AI-built sites most often fail silently. If search visibility is the problem, the cause is usually here.

Access: layers 1–4

1 · Technical access

  • Every important page returns a 200 response, not a redirect chain or a soft error.
  • robots.txt exists and does not Disallow: / the whole site.
  • No stray <meta name="robots" content="noindex"> left over from a template or staging setup.
  • The site is reachable on one canonical host (pick www or bare, redirect the other) over HTTPS.

2 · Rendering

  • Your real headline and body text appear in the initial HTML a crawler receives — not only after JavaScript executes.
  • If the site is a single-page app, it is server-side rendered or pre-rendered so content is present on first response.
  • You have tested a key page with JavaScript disabled and confirmed the content is still there.
# Fetch a page the way a crawler first sees it
curl -sL https://yoursite.com/ | grep -i "<h1"
# Empty result? Your content isn't in the initial HTML.

3 · Indexation

  • Each page has a self-referential <link rel="canonical"> pointing at its own real URL — never a hardcoded or preview-domain URL.
  • A sitemap.xml lists your indexable URLs and is referenced from robots.txt.
  • Pages you don’t want indexed (staging, thin utility pages) are deliberately excluded, not accidentally included.

4 · Architecture

  • URLs are clean, stable, and readable — not query-string state or hash routes.
  • Important pages are within a few clicks of the home page.
  • There are no orphaned pages — every indexable page is linked from somewhere crawlable.

Meaning: layers 5–8

5 · Metadata

  • Every page has a unique, specific <title> — not one shared default across the whole site.
  • Every page has a unique meta description written for a human scanning results.
  • Open Graph title and description are set so shared links render properly.

6 · Content

  • Each page answers a real question specifically, rather than restating generic category boilerplate.
  • Commercial pages describe what you actually do, for whom, and what happens next.
  • No large sets of near-duplicate, thin, or auto-generated pages that dilute the site.

7 · Internal links

  • Navigation and in-content links are real <a href> elements, not JavaScript click-handlers.
  • Related pages link to one another with descriptive anchor text.
  • Your most important commercial pages receive the most internal links.

8 · Structured data

  • Where it genuinely applies — organization, articles, products, FAQs — entities are described in JSON-LD.
  • Structured data reflects what is actually on the page; it never claims things the page doesn’t show.
<script type="application/ld+json">
{
  "@context": "https://schema.org",
  "@type": "Organization",
  "name": "Your Business",
  "url": "https://yoursite.com"
}
</script>

Competition: layers 9–12

9 · Performance

  • Pages load quickly and become visually stable without large layout shifts.
  • Images are sized and compressed; the JavaScript bundle isn’t doing work the server could do once.
  • Performance is measured with a real tool, not asserted — and never confused with the narrow response checks a scanner does.

10 · Authority

  • At least a few credible, relevant sites reference yours; a brand-new domain with none will rank slowly regardless of technical quality.
  • Authority is earned through genuine references and mentions — never bought link schemes, which are a liability.

11 · Conversion

  • Each page that ranks has a clear next step — the traffic is worthless if it bounces.
  • Contact and calls-to-action are present, honest, and functional.

12 · Measurement

  • Google Search Console is verified so you can see impressions, coverage, and which pages are actually indexed.
  • Analytics is in place to connect search traffic to real outcomes.
  • You check coverage and rankings on a cadence, rather than assuming a one-time fix holds.

Working through this in order is exactly what Search Foundation does for the access and indexation layers, and what Search Growth does for content, authority, and conversion over time. To see where your own site stands on the access layers right now, start with the free visibility scan.

Keep reading

Related reading and next steps

See it on your own site

Run a free visibility scan

This article explains what search engines need. The scan shows you which of those things your own public page actually delivers today — no score, no email gate.

Run a free visibility scan

This is a narrow technical scan of the public page response. It does not measure rankings, content quality, backlinks, Google index coverage, or business opportunity.