Applications open, taking on a few SaaS clients for Q3 2026. Engagements from $5,000/mo. See how it works
Playbooks

SEO audit checklist: 34 checks

Most audit checklists were written before AI search existed and stop at speed and on-page. This one adds the layer that decides whether AI engines can read and cite you at all, and orders everything by dependency.

By Mohammad Qaiser 28 July 202610 min read
An SEO audit run in dependency order, from crawling and indexing through to content

Most SEO audit checklists were written before AI search existed, and it shows. They cover crawling, on-page and speed thoroughly, then stop, which leaves out the layer that increasingly decides whether buyers ever see you at all.

The direct answer

An SEO audit checklist should cover five areas: crawling and indexing, on-page fundamentals, performance and mobile, content and intent, and AI visibility. The first four are standard. The fifth is missing from almost every checklist published before 2025, and it is the one that decides whether AI engines can read and cite you.

Thirty-four checks below, in dependency order, all tickable. Nothing is saved anywhere, so refreshing resets it. Work top down and stop wherever you run out of time; the ordering is deliberate.

Why the order matters

Checklists are usually grouped by category, which is tidy and unhelpful, because it implies the items are independent. They are not.

If your pages are not being rendered, every on-page fix below is wasted effort you will not detect for months. If your commercial pages have no conversion path, ranking them produces traffic and no revenue. Sequence encodes dependency, and working through a category-sorted list in arbitrary order is how audits produce eighty findings and no improvement.

An audit that produces eighty findings and no prioritization has moved the problem, not solved it.

One more thing before you start: an audit is only useful if someone can act on it. Before you begin, know who can approve technical changes and how fixes get into a sprint. Findings that sit in a backlog for a quarter are indistinguishable from findings nobody made.

Crawling and indexing

If search engines cannot reach or render a page, nothing below this section matters.

Crawling and indexing0 / 7
The crawl-to-index sequence from robots and sitemap through to pages actually indexed

On-page fundamentals

Unglamorous, cheap to fix, and still where a surprising share of easy wins sit.

On-page fundamentals0 / 7

Performance and mobile

Modest ranking impact, substantial conversion impact.

Performance and mobile0 / 6
Core Web Vitals pass marks, LCP under 2.5s, INP under 200ms, CLS under 0.1

Content and intent

Where audits usually go shallow, because it needs judgment rather than a crawler.

Content and intent0 / 6

AI visibility

The section missing from every classic audit checklist, and increasingly the one that decides shortlists.

AI visibility0 / 8
The five sections of the SEO audit checklist, how many checks each carries and what each section is looking for
SectionChecksWhat it is looking for
Crawling and indexing7Whether engines can reach and keep the pages that matter.
On-page fundamentals7Whether each page states clearly what it is and what it claims.
Performance and mobile6Whether speed or layout is costing conversions as well as rankings.
Content and intent6Whether the page answers the query someone actually typed.
AI visibility8Whether answer engines can extract and cite the page.
Total34Ordered so a finding never depends on a check you have not run yet.

What audit tools cannot tell you

Crawlers are excellent and they have blind spots. These are the checks that need a person, and they are disproportionately the ones that matter.

  • Whether content is rendered. Most crawlers fetch raw HTML by default, so a client-rendered page can pass a crawl and still be invisible to AI crawlers.
  • Whether a page matches search intent. A tool can tell you a page ranks. It cannot tell you it is answering a different question than the one being asked.
  • Whether a fix is worth the engineering cost. Severity scores are generic. Whether two sprints of developer time is justified depends on your roadmap.
  • Whether AI engines describe you accurately. No crawler checks this. You have to ask the engines directly.
  • Which of eighty findings actually matter. This is the judgment you are paying for when you hire someone, and the thing a scored report cannot supply.

A related caution: asking an AI assistant to audit a URL produces a plausible-looking list that misses most critical errors, because it cannot crawl properly, compare rendered against raw HTML, or read your Search Console data. Use a crawler for the audit and a model to help interpret the output. That split works well; the reverse does not.

If you would rather not run it yourself

This checklist is the first two weeks of every engagement

We run all thirty-four checks, plus log file analysis and a full AI citation baseline across seven engines, then deliver a fix list ranked by revenue impact rather than by severity score. The ranking is the deliverable. Anyone can produce eighty findings.

34
checks, in dependency order
7
AI engines baselined
2 to 3 weeks
for a full audit
$4k to $8k
as a standalone project, credited if you continue
  • Prioritized by revenue impact, with effort estimates your engineers can plan against
  • Implementation specs, not just findings. Written so a developer can act without interpreting
  • AI visibility included, which almost no audit currently covers
  • Credited against your first month if you go on to a retainer
Findings a crawler can produce set against findings only a person reading pages can

What to do with the findings

The audit is the easy part. Converting it into shipped changes is where most of them die.

  1. Sort by revenue impact, not severity. A critical-severity issue on a page nobody visits matters less than a moderate issue on your pricing page.
  2. Estimate effort alongside impact. The cheap high-value fixes should be identifiable at a glance, because those are what actually get done.
  3. Separate what you can ship from what needs engineering. Two different queues with two different owners.
  4. Fix the indexation blockers first, regardless of where they sit on the impact list. Everything downstream depends on them.
  5. Re-run the audit after the deploy. Rendering and canonical regressions ship silently with any front-end change.
  6. Set a recurring date. Quarterly for most sites, monthly if you ship weekly.

If you want the software-specific version with fourteen more checks around JavaScript rendering, programmatic pages and documentation cannibalization, the 47-point SaaS SEO checklist covers it. For the AI visibility section in depth, see LLM SEO.

A quadrant for triaging audit findings by impact against the effort each one takes

Frequently asked questions

What should an SEO audit checklist include?
Five areas: crawling and indexing, on-page fundamentals, performance and mobile, content and intent, and AI visibility. The first four are standard across published checklists. The fifth, covering Bing indexation, whether AI crawlers can read your pages, and how engines describe your brand, is missing from almost every checklist written before 2025.
How often should you run an SEO audit?
Quarterly for most sites, and monthly if you deploy frequently, because rendering and canonical regressions ship silently with front-end changes. Run a targeted re-audit after any migration, redesign or framework upgrade, and check indexation coverage continuously rather than waiting for the next full pass.
How long does an SEO audit take?
A thorough audit takes two to three weeks: crawling, comparing rendered against raw HTML, reviewing log files where available, checking indexation coverage, testing AI visibility and then prioritizing by revenue impact. A quick pass over the highest-impact checks can be done in a day, and is worth considerably more than nothing.
Can I do an SEO audit myself?
Yes, for most of it. The checklist above needs a crawler, Search Console access and about a day. Where outside help earns its fee is prioritization, deciding whether a fix justifies the engineering cost, and the technical diagnosis around rendering. Those need judgment rather than tooling.
What is the difference between a technical SEO audit and a full SEO audit?
A technical audit covers crawling, indexing, rendering, site architecture and performance. A full audit adds content and intent analysis, internal linking, competitive gaps and, increasingly, AI visibility. Technical is the prerequisite: content work published onto a site search engines cannot render produces nothing.
Can ChatGPT run an SEO audit for me?
Not reliably. It cannot crawl your site, compare rendered against raw HTML, read log files or access Search Console data, and testing consistently shows it misses a large share of obvious critical errors. Use a crawler to produce the findings, then use a model to help interpret and prioritize them, which it does well.
What are the most common SEO audit findings?
Pages excluded from the index for reasons nobody has read, canonical errors inherited from a template, content that only exists after JavaScript runs, duplicate or missing title tags, orphaned pages, crawl budget spent on filtered and parameter URLs, and commercial pages with no conversion path. Most are cheap to fix once identified.
Mohammad Qaiser

Mohammad Qaiser

Founder & Campaign Lead, Authority Magnet

Working in SEO since 2010, founder of Authority Magnet since 2018, and campaign lead on every case study the agency publishes. Also built PRWiz, where this work gets tested on our own software before it reaches a client.

Work with us

Want this run on your site?

Everything we publish is what we do for clients. Tell us the product and the category, and we will tell you what is realistic.

From $5,000/mo · Seven engines tracked · Reply within 24h